CoChat

The DOI matches its Crossref record; 34 of the 34 references it cites were found in a registry or an index; 12 of the 42 sources its claims cite were checked for support.

Energy-Saving Control Method for Factory Mushroom Room Air Conditioning Based on MPC

Open access

Abstract

The energy consumption of the mushroom room air conditioning system accounts for 40% of the total energy consumption of the mushroom factory. Efficient and energy-efficient mushroom factories and mushroom houses are the development direction of the industry. Compared with maintenance structure transformation and air conditioning equipment upgrading, energy-saving technology based on regulation methods has the advantages of less investment and fast effectiveness, which has attracted attention. The current methods for regulating air conditioning in edible mushroom factories include simple on/off thermostat control or PID. In the field of energy efficiency in commercial building air conditioning, a large number of studies have shown that compared with traditional control algorithms such as classic on/off or PID control, model predictive control can significantly improve energy efficiency. However, there is little literature mentioning the application of MPC in factory mushroom production rooms. This paper proposes a data-driven MPC and PID combined energy-saving control method for mushroom room air conditioning. This method uses the CNN-GRU-Attention combination neural network as the prediction model, combined with prediction error compensation and dynamic update mechanism of the prediction model dataset, to achieve an accurate prediction of indoor temperature in mushroom houses. Establish an objective function for air conditioning control duration and temperature, use the non-dominated sorting genetic algorithm II (NSGA-II) to solve for the optimal control sequence of the air conditioning in the control time domain, and use the entropy weight method to determine the optimal decision quantity. Integrate rolling optimization, feedback mechanism, and PID to achieve precise and energy-saving control of the mushroom room environment. The experimental results show that compared with the on/off thermostat and PID controller, the designed controller reduces power consumption by 12% and 5%, respectively, and has good application and demonstration value in the field of industrial production of edible mushrooms.

Overview

CoChat Summary

TL;DR

This paper proposes a data-driven MPC combined with PID for factory mushroom room air conditioning, using a CNN-GRU-Attention prediction model, NSGA-II optimization, and entropy-weight decision-making. In experiments, the designed controller reduced power consumption by 12% and 5% compared with an on/off thermostat and a PID controller, respectively.

Key findings

  • The Boruta feature-selection method improved prediction over Pearson correlation: R² = 0.8912 and RMSE = 0.1434 °C, versus R² = 0.8791 and RMSE = 0.1557 °C for Pearson. A time step of 3 was selected, using historical 30 min data to predict the next 10 min indoor temperature.
  • A 30 min control horizon was chosen. In the control-horizon comparison, air-conditioning duration was 717 min at 0–10 min, 696 min at 0–20 min, 686 min at 0–30 min, 681 min at 0–40 min, and 683 min at 0–50 min; optimization duration rose from 43 s to 156 s.
  • On July 16, the designed controller kept indoor temperature between 14.6403 and 14.3823 °C (median 14.5056 °C), compared with PID between 14.8376 and 14.1614 °C (median 14.4994 °C) and on/off between 15.2831 and 13.5876 °C (median 14.2267 °C).
  • Daily air-conditioning opening time was 701 min for the designed controller, 741 min for PID, and 797 min for the on/off thermostat. The paper reports that the designed controller reduced power consumption by 12% and 5% versus on/off and PID, respectively, and reduced temperature-control RMSE by more than 90% versus both.

Methods

  • Case study in an edible mushroom company in Beijing producing *Hypsizygus marmoreus*, using three mushroom rooms of 112 m² and about 3 m height with constant-frequency air conditioning. Data were collected from 12 May 2022 to 12 September 2022 with sensors sampling at 1 min and averaged over 10 min.
  • The controller uses CNN-GRU-Attention as the MPC prediction model, Boruta feature selection, prediction-error compensation, and dynamic dataset updating (3312 pieces per 23 days, with 144 new pieces per day). The objective function minimizes temperature error and air-conditioning duration; NSGA-II solves the Pareto set and entropy weight selects the decision.
  • A two-level scheme is used: Level 1 MPC re-optimizes the air-conditioning control quantity every 10 min, and Level 2 PID regulates indoor temperature with a 2 min control cycle, constrained by the MPC-determined total duration. Performance is evaluated with RMSE and R², feature-selection comparisons, time-step tests, control-horizon tests, and comparison against on/off and PID controllers.

Limitations

  • The paper states that it does not consider the energy consumption of the pre-cooling device. It also states future work will explore feature decomposition methods such as empirical wavelet transform and incorporate the heat balance equation into the prediction model loss function to improve accuracy and robustness.
  • Evident from the text: the experiment is limited to one company, one mushroom type, three rooms, and one summer period (12 May–12 September 2022), so generalization to other sites, seasons, or mushroom species is not established.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/en16227623

Cites

34 of 34 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Schematic diagram of sensor installation in mushroom room. Figure 1. Schematic diagram of sensor installation in mushroom room.
Figure 1 Schematic diagram of sensor installation in mushroom room. Figure 1. Schematic diagram of sensor installation in mushroom room.From the PDF
Figure 2 Schematic diagram of the controller's principle. Figure 2. Schematic diagram of the controller's principle.
Figure 2 Schematic diagram of the controller's principle. Figure 2. Schematic diagram of the controller's principle.From the PDF
Figure 3 Structure diagram of the prediction model.
Figure 3 Structure diagram of the prediction model.From the PDF
Figure 4 Performance evaluation of the prediction models under different feature sets. ( a ) The correlation analysis between the predicted target and the predicted data in the dataset selected using the Pearson correlation coefficient method. ( b ) The error histogram between the predicted target and the predicted data in the dataset selected using the Pearson correlation coefficient method. ( c ) The changes in model loss value in the training and validation sets selected using the Pearson correlation coefficient method. ( d ) The correlation analysis between the predicted target and the predicted data in the dataset selected using the Boruta method. ( e ) The error histogram between the predicted target and the predicted data in the dataset selected using the Boruta method. ( f ) The changes in model loss value in the training and validation sets selected using the Boruta method. The R 2 value of the Pearson correlation coefficient method was 0.8791. The R 2 value Figure 4. Performance evaluation of the prediction models under different feature sets. ( a ) The correlation analysis between the predicted target and the predicted data in the dataset selected using the Pearson correlation coefficient method. ( b ) The error histogram between the predicted target and the predicted data in the dataset selected using the Pearson correlation coefficient method. ( c ) The changes in model loss value in the training and validation sets selected using the Pearson correlation coefficient method. ( d ) The correlation analysis between the predicted target and the predicted data in the dataset selected using the Boruta method. ( e ) The error histogram between the predicted target and the predicted data in the dataset selected using the Boruta method. ( f ) The changes in model loss value in the training and validation sets selected using the Boruta method.
Figure 4 Performance evaluation of the prediction models under different feature sets. ( a ) The correlation analysis between the predicted target and the predicted data in the dataset selected using the Pearson correlation coefficient method. ( b ) The error histogram between the predicted target and the predicted data in the dataset selected using the Pearson correlation coefficient method. ( c ) The changes in model loss value in the training and validation sets selected using the Pearson correlation coefficient method. ( d ) The correlation analysis between the predicted target and the predicted data in the dataset selected using the Boruta method. ( e ) The error histogram between the predicted target and the predicted data in the dataset selected using the Boruta method. ( f ) The changes in model loss value in the training and validation sets selected using the Boruta method. The R 2 value of the Pearson correlation coefficient method was 0.8791. The R 2 value Figure 4. Performance evaluation of the prediction models under different feature sets. ( a ) The correlation analysis between the predicted target and the predicted data in the dataset selected using the Pearson correlation coefficient method. ( b ) The error histogram between the predicted target and the predicted data in the dataset selected using the Pearson correlation coefficient method. ( c ) The changes in model loss value in the training and validation sets selected using the Pearson correlation coefficient method. ( d ) The correlation analysis between the predicted target and the predicted data in the dataset selected using the Boruta method. ( e ) The error histogram between the predicted target and the predicted data in the dataset selected using the Boruta method. ( f ) The changes in model loss value in the training and validation sets selected using the Boruta method.From the PDF
Figure 5 R 2 , RMSE , and training time of model at different time steps. ( a ) The RMSE and training time of the model at different time steps. ( b ) The R 2 at different time steps.
Figure 5 R 2 , RMSE , and training time of model at different time steps. ( a ) The RMSE and training time of the model at different time steps. ( b ) The R 2 at different time steps.From the PDF
Figure 6 Comparison diagram of the on/off thermostat, PID, and designed controller control temperature. 16 , x FOR PEER REVIEW 14 of 17
Figure 6 Comparison diagram of the on/off thermostat, PID, and designed controller control temperature. 16 , x FOR PEER REVIEW 14 of 17From the PDF
Image 7
From the PDF

Verification

Claims the paper makes

8 Valid · 3 Not valid · 18 Not checked · 35 Own results and interpretations

18 more claims

Create a free account to see all 64 claims and every source verdict.

See all claims — free account
  • Probably not a substantial claim. 27
  • It may state two separate claims as one. 24
  • Borderline: one of the tests for a substantial claim is unclear. 1

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Show all 32 references

Research works printed but not linked: 1

  1. Guan, 2010, Energy saving analysis of industrial production of edible mushrooms, Edible Fungi, 32, 1

Cited by

See all 8

Metadata

Authors
Wang, M., Zheng, W., Zhao, C., Chen, Y., Chen, C., & Zhang, X.
Affiliations
  • Shenyang Agricultural University (CN)
  • Beijing Academy of Agriculture and Forestry Sciences (CN)
Corresponding author
Zhao, Chunjiang
Venue
Energies
Publisher
MDPI AG
Published
2023-11-17
Volume
16
Issue
22
Pages
7623
ISSN
1996-1073 (ISSN-L)
OpenAlex
W4388762107
Cited by
8
References
32
Citations by year
2026: 1 · 2025: 4 · 2024: 3
FWCI
0.96
Type
journal article
Language
en
Open access
Locations
  • Energies (publishedVersion, cc-by)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
Beijing Academy of Agricultural and Forestry Sciences Science and Technology Innovation Capacity Construction Project (KJCX20230410)
Topics
  • Physical Sciences › Engineering › Building and Construction › Building Energy and Comfort Optimization
  • Physical Sciences › Engineering › Control and Systems Engineering › Advanced Control Systems Optimization
  • Physical Sciences › Engineering › Mechanical Engineering › Refrigeration and Air Conditioning Technologies
Keywords
energy-efficient control, model predictive control, edible mushroom cultivation, indoor temperature prediction, prediction error compensation, NSGA-II, entropy weight method, PID control, air conditioning energy consumption, data-driven control

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-09-28, under version 3 of the claim definition (prompt extract-v3).

A second model labelled each claim on 2026-09-28 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Energy-Saving Control Method for Factory Mushroom Room Air Conditioning Based on MPC. https://app.cochat.ai/papers/10.3390/en16227623
The paper
Wang, M., Zheng, W., Zhao, C., Chen, Y., Chen, C., & Zhang, X. (2023). Energy-Saving Control Method for Factory Mushroom Room Air Conditioning Based on MPC. Energies. https://doi.org/10.3390/en16227623

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.