Student or educator? Sign up with your school email and get 6 months of Pro free.

The DOI matches its Crossref record; 60 of the 60 references it cites were found in a registry or an index; 5 of the 20 sources its claims cite were checked for support.

Prediction of Heat and Cold Loads of Factory Mushroom Houses Based on EWT Decomposition

Open access

Abstract

Load forecasting has significant implications on optimizing the operation of air conditioning systems for industrial mushroom houses and energy saving. This research paper presents a novel approach for short-term load forecasting in mushroom houses, which face challenges in accurately modeling cold and heat loads due to the complex interplay of various factors, including climatic conditions, mushroom growth, and equipment operation. The proposed method combines empirical wavelet transform (EWT), hybrid autoregressive integrated moving average (ARIMA), convolutional neural network (CNN), and bi-directional long short-term memory (BiLSTM) with an attention mechanism (CNN-BiLSTM-Attention) to address these challenges. The first step of this method was to select input features via the Boruta algorithm. Then, the EWT method was used to decompose the load data of mushroom houses into four modal components. Subsequently, the Lempel–Ziv method was introduced to classify the modal components into high-frequency and low-frequency classes. CNN-BiLSTM-Attention and ARIMA prediction models were constructed for these two classes, respectively. Finally, the predictions from both classes were combined and reconstructed to obtain the final load forecasting value. The experimental results show that the Boruta algorithm selects key influential feature factors effectively. Compared to the Spearman and Pearson correlation coefficient methods, the mean absolute error (MAE) of the prediction results is reduced by 14.72% and 3.75%, respectively. Compared to the ensemble empirical mode decomposition (EEMD) method, the EWT method can reduce the decomposition reconstruction error by an order of magnitude of 103, effectively improving the accuracy of the prediction model. The proposed model in this paper exhibits significant advantages in prediction performance compared to the single neural network model, with the MAE, root mean square error (RMSE), and mean absolute percentage error (MAPE) of the prediction results reduced by 31.06%, 26.52%, and 39.27%, respectively.

Overview

CoChat Summary

TL;DR

This paper proposes a short-term heat/cold load forecasting method for factory mushroom houses that combines EWT decomposition, Lempel-Ziv component classification, Boruta feature selection, and separate ARIMA and CNN-BiLSTM-Attention models for low- and high-frequency components. Using data from a Beijing *Hypsizygus marmoreus* factory, the hybrid model reports lower MAE, RMSE, and MAPE than several comparison models.

Key findings

  • Boruta feature selection accepted 7 of 9 input features and rejected 2; the resulting model had MAE 0.591 kW, RMSE 1.056 kW, and MAPE 5.68%, with MAE reduced by 16.88%, 14.72%, and 3.75% compared with the control, Spearman, and Pearson methods, respectively, and training time 2 s shorter than Spearman/Pearson.
  • EWT decomposition at scale 4 gave the smallest reconstruction error (sum of absolute errors 0.05737 kW in Table 4); EWT reconstruction error fluctuated within ±3.0 × 10⁻⁴ versus ±8 × 10⁻¹ for EEMD, a difference of order 10³.
  • The proposed hybrid ARIMA + CNN-BiLSTM-Attention model based on EWT reduced MAE, RMSE, and MAPE by 31.06%, 26.52%, and 39.27%, respectively, compared with CNN-BiLSTM-Attention, and achieved R² = 0.951.
  • NSGA-II hyperparameter optimization gave MAE 0.557 kW, RMSE 0.884 kW, and MAPE 5.26%, with MAE 4.46% and 2.28% lower than PSO and GA, respectively; the segmented loss function gave MAE 0.497 kW and MAPE 4.64%.

Methods

  • Data were collected from an experimental mushroom house in Tongzhou District, Beijing, producing *Hypsizygus marmoreus*; the house was 14 m × 8 m × 5 m with polyurethane steel plate enclosures. Sensors measured indoor/outdoor temperature and humidity, air conditioner current, fresh air velocity, CO₂ concentration, solar radiation, and wall heat flux, with 1-min sampling from 11 July 2022 to 15 September 2022, totaling 729,300 data.
  • Preprocessing used linear interpolation for missing data and mean smoothing for abnormal data, followed by Boruta feature selection and normalization. EWT decomposed the load into four modal components; Lempel-Ziv complexity classified the first three as high-frequency and the fourth as low-frequency. CNN-BiLSTM-Attention predicted the high-frequency component, ARIMA predicted the low-frequency component, and the results were reconstructed.
  • The high-frequency model used historical 4-h load data to predict the next 10 min, with input feature dimension (24,7); the dataset was split 7:2:1 for training, validation, and testing. ARIMA parameters were p = 2, d = 1, q = 1. Hyperparameters were optimized with NSGA-II, and a segmented loss function was used; evaluation metrics were RMSE, MAE, MAPE, and R².

Limitations

  • The paper states that further improvements should leverage technologies such as online sensors and data cloud platforms to ensure continuous optimization of the model.
  • Evident from the text: validation used a single experimental mushroom house, one mushroom species (*Hypsizygus marmoreus*), and one summer period (11 July–15 September 2022) in Beijing; no multi-site, multi-season, or multi-species validation is reported. The paper does not present a formal limitations section.

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

Integrity signals

Registry

Crossref · 10.3390/su152115270

Cites

60 of 60 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

5 Valid · 1 Not valid · 12 Not checked · 94 Own results and interpretations

72 more claims

Create a free account to see all 112 claims and to verify the 4 references still open.

See all claims — free account
  • It may state two separate claims as one. 28
  • Probably not a substantial claim. 7
  • The quote only partly states this claim. 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.

3 of 3

Citations

Cites

Show all 44 references

Research works printed but not linked: 3

  1. Guan, 2010, Energy-saving analysis on industrial production of edible mushrooms, Edible Fungi, 32, 1
  2. Kapalo, P., Voznyak, O., Zhelykh, V., and Klymenko, H. (2023). Proceedings of EcoComfort 2022, Springer. Lecture Notes in Civil Engineering, 290 LNCE.
  3. Kapalo, P., Sulewska, M., and Adamski, M. (2021). Proceedings of EcoComfort 2020, Springer. Lecture Notes in Civil Engineering, 100 LNCE.

Cited by

See all 8

Metadata

Authors
Zuo, H., Zheng, W., Wang, M., & Zhang, X.
Affiliations
  • Guangxi University (CN)
  • Beijing Academy of Agriculture and Forestry Sciences (CN)
Corresponding author
Wang, Mingfei
Venue
Sustainability
Publisher
MDPI AG
Published
2023-10-25
Volume
15
Issue
21
Pages
15270
ISSN
2071-1050 (ISSN-L)
OpenAlex
W4388263247
Cited by
8
References
44
Citations by year
2026: 3 · 2025: 3 · 2024: 2
FWCI
0.60
Type
journal article
Language
en
Open access
Locations
  • Sustainability (publishedVersion, cc-by)
  • Preprints.org (acceptedVersion, cc-by)
  • RePEc: Research Papers in Economics (submittedVersion)
Funders
  • National Edible Mushroom Industry Technology System (CARS-20, BAIC03-2023)
  • Beijing Edible Mushroom Innovation Team (CARS-20, BAIC03-2023)
Topics
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Energy Load and Power Forecasting
  • Physical Sciences › Engineering › Building and Construction › Building Energy and Comfort Optimization
  • Life Sciences › Agricultural and Biological Sciences › Plant Science › Smart Agriculture and AI
Keywords
short-term load forecasting, mushroom house, heat load, cold load, empirical wavelet transform, CNN-BiLSTM-Attention, ARIMA, Boruta feature selection, air conditioning energy saving

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: Prediction of Heat and Cold Loads of Factory Mushroom Houses Based on EWT Decomposition. https://app.cochat.ai/papers/10.3390/su152115270
The paper
Zuo, H., Zheng, W., Wang, M., & Zhang, X. (2023). Prediction of Heat and Cold Loads of Factory Mushroom Houses Based on EWT Decomposition. Sustainability. https://doi.org/10.3390/su152115270

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.