Automatic waste heat recovery method

Abstract

The invention provides an automatic waste heat recovery method which comprises the steps that waste heat pressure parameters, component media and flow characteristic data are obtained, waste heat data are classified, characteristic vectors of different types of waste heat are obtained, and the characteristic vectors of different types of preheating are stored in a waste heat characteristic database; analyzing and predicting randomness and intermittency rules of waste heat generation and a waste heat concentration area, and dynamically adjusting a cold absorption strategy or a heat absorption strategy; corresponding cold absorption or heat absorption operation parameters are called from the waste heat characteristic database, and cold absorption or heat absorption equipment is controlled to set a self-adaptive adjustment operation state according to the parameters; the cold and heat storage energy in the recovery process is monitored in real time, and the energy loss value of waste heat recovery is analyzed; and constructing a photovoltaic power storage energy optimization scheduling model, solving to obtain an energy distribution strategy, dynamically adjusting a use strategy of cold and heat storage energy, optimizing a distribution proportion and a time sequence of cold and heat energy, and realizing collaborative optimization control of photovoltaic power storage and waste heat recovery.

Description

Technical Field

[0001] The invention relates to the field of information technology, and in particular to an automatic waste heat recovery method. Background Art

[0002] If the waste heat generated by the air conditioner in the office environment or the air compressor outside the company is not recovered, it will lead to a waste of energy resources. Not only that, with the rise in energy prices, the pressure on enterprises in operating costs continues to increase, and the failure to recover waste heat will directly affect the economic benefits of the enterprise. Through waste heat recovery, the waste heat that was originally wasted can be converted into usable energy, thereby reducing the demand for heating and cooling, reducing energy costs, and improving the economic benefits of the enterprise. Therefore, when automatically recovering waste heat in the office scene, there is a technical problem of how to accurately analyze and process it according to the pressure, component medium and flow characteristics of different waste heat data. Due to the diverse and dispersed sources of waste heat in the office environment, the pressure, component medium and flow parameters of various types of waste heat vary greatly, and the waste heat content is relatively small, which makes it difficult for the system to uniformly absorb cold and heat. In addition, the generation of different waste heat is random and intermittent. How to achieve real-time response and dynamic adjustment of the system to match the rhythm of waste heat generation is also a major challenge. At the same time, in the process of absorbing cold and heat from waste heat, how to avoid energy loss and efficiency reduction, and ensure the quality and quantity of cold and heat storage, also requires breaking through technical bottlenecks. Although photovoltaic power storage can provide energy for the system, how to optimize the energy scheduling and allocation strategy of photovoltaic power storage according to the actual needs of waste heat recovery and achieve efficient operation and adaptive control of the system still needs further research. Summary of the invention

[0003] The present invention provides an automatic waste heat recovery method, which mainly includes: Obtain waste heat pressure parameters, component medium and flow characteristic data, classify waste heat data, obtain characteristic vectors of different types of waste heat, including temperature, pressure, flow, and heat capacity parameters, and store characteristic vectors of different types of preheating in a waste heat characteristic database; Obtain the waste heat distribution inside and outside the office environment, identify the concentrated areas of each waste heat source, collect the generation data of each waste heat source in real time, analyze and predict the randomness and intermittent laws of waste heat generation and the waste heat concentrated areas, and dynamically adjust the cooling or heat absorption strategy; Retrieving corresponding cold absorption or heat absorption operation parameters from the waste heat characteristic database, and controlling the cold absorption or heat absorption equipment to adaptively adjust the operation state according to the parameter settings; Real-time monitoring of the cold and heat storage energy during the recovery process is carried out, and the energy loss value of the remaining heat recovery is analyzed. If the energy loss exceeds the preset threshold, the fault diagnosis program is triggered to determine the fault type and give the corresponding optimization control strategy; Construct a photovoltaic power storage energy optimization scheduling model, solve the energy allocation strategy, dynamically adjust the use strategy of cold and heat storage energy, optimize the distribution ratio and timing of cold and heat energy, and realize the coordinated optimization control of photovoltaic power storage and waste heat recovery.

[0004] Furthermore, the acquisition of waste heat pressure parameters, component medium and flow characteristic data includes: The temperature parameters, pressure characteristics, flow data and heat capacity properties of the waste heat recovery system are obtained, and the numerical range of the parameter data is checked. If the parameter value exceeds the preset threshold range, it is marked as abnormal data; the abnormal data is linearly interpolated or eliminated, and the normal data is normalized to obtain standardized parameter values; the waste heat medium is analyzed by gas chromatograph to identify the main chemical components and their contents, and the component data is merged with the standardized parameter value.

[0005] Furthermore, the classification of the waste heat data includes: The K-means clustering algorithm is used to classify the merged multidimensional data to obtain feature clusters of different types of waste heat, and the centroid coordinates of each feature cluster are calculated as the feature vector of the waste heat; a waste heat feature database table is established, which includes waste heat type, temperature vector, pressure vector, flow vector, heat capacity vector and component vector, and the feature vectors of different types of waste heat are inserted into the waste heat feature database table for storage.

[0006] Furthermore, obtaining the waste heat distribution inside and outside the office environment includes: Obtaining an ambient temperature distribution map, wherein the ambient temperature distribution map is obtained by acquiring real-time data of multiple temperature sensors around the exhaust port of the air compressor outside the office environment and the air conditioner condenser inside the office environment, and estimating the waste heat distribution inside and outside the office environment using the Kriging interpolation algorithm according to the exhaust temperature and exhaust flow rate of the air compressor and the exhaust temperature and exhaust flow rate parameters of the air conditioner condenser; A K-means clustering algorithm is executed according to the ambient temperature distribution map to obtain an abnormal temperature area as a waste heat source concentration area; a sensor network is arranged around the waste heat source concentration area, and the sensor network includes a temperature sensor, a flow sensor and a pressure sensor.

[0007] Furthermore, the analysis predicts the randomness and intermittent regularity of waste heat generation and waste heat concentration areas, including: Collect waste heat source data acquired by the sensor network, the waste heat source data including temperature data, flow data and pressure data; perform time series analysis on the waste heat source data, the time series analysis including data preprocessing, trend analysis and seasonal analysis; use an autoregressive moving average model to process the waste heat source data to obtain characteristic curves and prediction models for each waste heat source.

[0008] Furthermore, the dynamic adjustment of the cold absorption strategy or the heat absorption strategy includes: The optimal cold absorption or heat absorption strategy is calculated according to the characteristic curve and the prediction model, and the optimal cold absorption or heat absorption strategy is obtained by a linear programming method; if the waste heat generation of the prediction model is greater than a preset threshold, the cold absorption device is started and the cooling power is adjusted; if the waste heat generation of the prediction model is less than the preset threshold, the heat absorption device is started and the heating power is adjusted.

[0009] Furthermore, the step of retrieving corresponding cold absorption or heat absorption operation parameters from the residual heat characteristic database includes: According to the acquired ambient temperature and humidity data, the waste heat feature record with the highest matching degree is retrieved from the waste heat feature database, and the waste heat feature database adopts a B-tree index structure; according to the waste heat feature record, initial cooling or heat absorption operation parameters are obtained, and the initial cooling or heat absorption operation parameters include a heat pump speed range and an expansion valve opening range; a fuzzy logic controller is used to optimize the initial cooling or heat absorption operation parameters, and the input variables of the fuzzy logic controller include temperature deviation and humidity deviation.

[0010] Furthermore, the control of the cold or heat absorbing device to adaptively adjust the operating state according to parameter settings includes: The output frequency of the heat pump inverter is adjusted in real time by a PID controller, and the PID controller adopts the Ziegler-Nichols method to perform PID parameter self-tuning; the temperature, pressure, and flow data of the condenser and the evaporator are obtained, and the temperature, pressure, and flow data are preprocessed by sliding average filtering; according to the preprocessed temperature, pressure, and flow data, the heat pump speed and expansion valve opening are fine-tuned in real time by using an adaptive neural fuzzy inference system, and the adaptive neural fuzzy inference system continuously optimizes the inference rules through online learning.

[0011] The technical solution provided by the embodiment of the present invention may have the following beneficial effects: The present invention discloses an automatic waste heat recovery method. By deploying an infrared thermal imager and a sensor network, the temperature, pressure, flow and other data of waste heat sources inside and outside the office area are obtained in real time, the distribution characteristics and generation rules of waste heat are analyzed, and the cold absorption or heat absorption strategy is dynamically adjusted. In combination with the parameters in the waste heat characteristic database, the operating state of the cold absorption or heat absorption equipment is controlled to achieve adaptive adjustment of the heat pump speed and the expansion valve opening. At the same time, an energy metering device is arranged in the waste heat transmission pipeline network to monitor the recovery process of cold storage and heat storage energy in real time. If the energy loss exceeds the preset threshold, the fault diagnosis program is triggered and the control strategy is optimized. In addition, the power generation data of the photovoltaic cell and the charge state of the battery are obtained, and the energy loss of waste heat recovery is combined to construct a photovoltaic power storage energy optimization scheduling model, dynamically adjust the use strategy of cold storage and heat storage energy, and realize the coordinated optimization control of photovoltaic power storage and waste heat recovery. By continuously monitoring the system operation data, the potential efficiency reduction or failure risk is timely identified to ensure the efficient operation of the system. The present invention realizes efficient management of office environment energy, reduces energy loss, and improves the overall energy efficiency of the system through the coordinated optimization of waste heat recovery and photovoltaic power storage. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 The present invention is a flow chart of an automatic waste heat recovery method.

[0013] Figure 2 It is a schematic diagram of an automatic waste heat recovery method of the present invention.

[0014] Figure 3 It is another schematic diagram of an automatic waste heat recovery method of the present invention. DETAILED DESCRIPTION

[0015] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0016] like Figure 1-3 In this embodiment, an automatic waste heat recovery method may specifically include: Step S101, obtain waste heat pressure parameters, component medium and flow characteristic data, classify the waste heat data, obtain characteristic vectors of different types of waste heat, including temperature, pressure, flow, and heat capacity parameters, and store the characteristic vectors of different types of preheating in a waste heat characteristic database.

[0017] The temperature parameters, pressure characteristics, flow data and heat capacity properties of the waste heat recovery system are obtained, and the parameter data are checked for a numerical range. If the parameter value exceeds the preset threshold range, it is marked as abnormal data; the abnormal data is linearly interpolated or eliminated, and the normal data is normalized to obtain a standardized parameter value; the waste heat medium is analyzed by gas chromatograph to identify the main chemical components and their contents, and the component data is merged with the standardized parameter value; the K-means clustering algorithm is used to classify the merged multidimensional data to obtain characteristic clusters of different types of waste heat, and the centroid coordinates of each characteristic cluster are calculated as the characteristic vector of the waste heat; the waste heat medium is analyzed by gas chromatograph to identify the main chemical components and their contents, and the component data is merged with the standardized parameter value; the K-means clustering algorithm is used to classify the merged multidimensional data to obtain characteristic clusters of different types of waste heat, and the centroid coordinates of each characteristic cluster are calculated as the characteristic vector of the waste heat; A thermal characteristic database table includes waste heat type, temperature vector, pressure vector, flow vector, heat capacity vector and component vector, and inserts characteristic vectors of different types of waste heat into the waste heat characteristic database table for storage; reads the stored characteristic vectors from the waste heat characteristic database table, performs principal component analysis by calculating the covariance matrix, eigenvalues ​​and eigenvectors, and selects the first few eigenvectors whose cumulative contribution rate exceeds a preset threshold as the main characteristic components; establishes a multivariate linear regression model based on the main characteristic components, uses the main characteristic components as independent variables, and waste heat utilization efficiency as a dependent variable, and calculates the regression coefficient to obtain the multivariate linear regression model of the waste heat characteristics.

[0018] Specifically, temperature parameters, pressure characteristics, flow data and heat capacity properties are obtained from the waste heat recovery system, and the obtained parameter data are checked for a numerical range. If the parameter value exceeds the preset threshold range, it is marked as abnormal data, and the abnormal data is linearly interpolated or eliminated. The normal data is normalized to obtain a standardized parameter value. The waste heat medium is analyzed by gas chromatograph to identify the main chemical components and their contents, and the component data is merged with the standardized parameter value. The K-means clustering algorithm is used to classify the merged multidimensional data to obtain feature clusters of different types of waste heat, and the centroid coordinates of each feature cluster are calculated as the feature vector of this type of waste heat. The centroid coordinates are obtained by calculating the average value of all data points in the cluster. A waste heat feature database table is established, which includes fields such as waste heat type, temperature vector, pressure vector, flow vector, heat capacity vector, and component vector. The feature vectors of different types of waste heat are inserted into the database table for storage. The stored feature vectors are read from the database, and principal component analysis is performed by calculating the covariance matrix, eigenvalues, and eigenvectors. The first few eigenvectors whose cumulative contribution rate exceeds the preset threshold are selected as the main feature components. A multivariate linear regression model is established based on the main characteristic components. The main characteristic components are used as independent variables, and the waste heat utilization efficiency is used as the dependent variable. The regression coefficient is calculated to obtain the mathematical model of the waste heat characteristics for subsequent waste heat classification and matching. Temperature, pressure, flow rate and heat capacity data are collected from the waste heat recovery system, such as temperature range 0-1000℃, pressure range 0-10MPa, flow rate range 0-1000m³ / h, and heat capacity range 0-5kJ / (kg·K). Anomaly detection is performed on the collected data, and the temperature threshold is set to 950℃. Data points exceeding this value are corrected by linear interpolation. Normal data is normalized to the range of 0-1. A gas chromatograph is used to analyze the composition of the waste heat medium, identify the main components such as CO2, H2O, N2, etc., and record the percentage of each component. The component data is combined with the normalization parameters to form a multidimensional feature vector. The K-means clustering algorithm is used to classify the feature vectors, and the number of clusters K=5 is set. Iterations are repeated 50 times to obtain stable clustering results. Calculate the centroid coordinates of each cluster, such as the centroid coordinates of cluster 1 (0.3, 0.5, 0.7, 0.4, 0.6) represent the eigenvector of this type of waste heat. Create a waste heat feature table in the database, including fields: ID, type, temperature, pressure, flow, heat capacity, CO2 content, H2O content, N2 content, etc. Store the calculated eigenvectors into the database, such as inserting records (1, 'high temperature waste heat', 0.3, 0.5, 0.7, 0.4, 0.6, 0.3, 0.1). Read all eigenvectors from the database and construct a sample matrix. Calculate the sample covariance matrix and solve the eigenvalues ​​and eigenvectors. Select the first three eigenvectors with a cumulative contribution rate of 95% as the principal components. Establish a multivariate linear regression model based on the principal components, with the independent variables being the scores of the three principal components and the dependent variable being the waste heat utilization efficiency.The regression coefficients were estimated by the least squares method, and the model equation was obtained: y=0.72x1+0.56x2+0.33x3+0.15, where y is the waste heat utilization efficiency, and x1, x2, and x3 are the three principal component scores. This model can be used to predict the utilization efficiency of different types of waste heat and assist in waste heat matching and classification.

[0019] Step S102, deploy infrared thermal imagers inside and outside the office area to obtain the waste heat distribution inside and outside the office environment and identify the concentrated areas of each waste heat source; and install a number of sensors to form a sensor network to collect the generation data of each waste heat source in real time. The generation data of each waste heat source includes the temperature, flow rate, and pressure of the waste heat source. According to the analysis of the generation data of each waste heat source, the randomness and intermittent law of waste heat generation and the waste heat concentration area are predicted, and the cooling strategy or the heat absorption strategy is dynamically adjusted.

[0020] Acquire an ambient temperature distribution map, the ambient temperature distribution map is obtained by acquiring real-time data of multiple temperature sensors around the exhaust port of the air compressor outside the office environment and the air conditioner condenser in the office environment, and using the Kriging interpolation algorithm to estimate the waste heat distribution inside and outside the office environment according to the exhaust temperature and exhaust flow rate of the air compressor and the exhaust temperature and exhaust flow rate parameters of the air conditioner condenser; execute the K-means clustering algorithm according to the ambient temperature distribution map to obtain the temperature abnormality area as the waste heat source concentration area; arrange a sensor network around the waste heat source concentration area, the sensor network includes temperature sensors, flow sensors and pressure sensors; collect the waste heat source data acquired by the sensor network, the waste heat source data includes temperature data, flow data and pressure data; perform time series analysis on the waste heat source data Analysis, the time series analysis includes data preprocessing, trend analysis and seasonal analysis; the waste heat source data is processed by an autoregressive moving average model to obtain a characteristic curve and a prediction model of each waste heat source; the optimal cooling or heat absorption strategy is calculated according to the characteristic curve and the prediction model, and the optimal cooling or heat absorption strategy is obtained by a linear programming method; if the waste heat generation of the prediction model is greater than a preset threshold, the cooling device is started and the cooling power is adjusted; if the waste heat generation of the prediction model is less than the preset threshold, the heat absorption device is started and the heating power is adjusted; the operation effect of the cooling device or the heat absorption device is monitored, and the operation effect is obtained through a sensor network; the equipment operation parameters are adjusted according to the deviation between the operation effect and the expected effect, and the equipment operation parameters include cooling power or heating power.

[0021] Specifically, infrared thermal imagers are fixedly installed inside and outside the office area, such as air conditioners in the office area or air compressors outside the office area, etc. The ambient temperature distribution map is obtained through thermal imaging technology, and the temperature distribution map is processed using the K-means clustering algorithm. The area with a temperature 20% higher than the ambient average temperature is identified as an abnormal temperature area as a waste heat source concentration area, and the spatial coordinates of each waste heat source are recorded. Temperature sensors, flow sensors and pressure sensors are arranged around the identified waste heat source concentration area, and a sensor network is constructed. The data collection frequency is set to once per minute, and the temperature, flow and pressure data of each waste heat source are collected in real time. The collected data are sent to the data processing center through a wireless transmission module. At the same time, data integrity checks and sensor self-test programs are set to ensure the reliability of data transmission. Time series analysis is performed on the waste heat source data collected by the sensor network, including data preprocessing, trend analysis and seasonal analysis. The autoregressive moving average model is used to calculate the periodicity and volatility of each waste heat source data. The lag period is set to 12 and the moving average period is set to 3. The randomness and intermittent laws of waste heat generation are judged, and the characteristic curve and prediction model of each waste heat source are obtained. According to the residual heat source characteristic curve and prediction model, combined with the current ambient temperature and humidity data, the optimal cooling or heat absorption strategy is calculated using the linear programming method. If the predicted residual heat generation is greater than the preset threshold, the cooling device is started and the cooling power is adjusted. If the predicted residual heat generation is less than the preset threshold, the heat absorption device is started and the heating power is adjusted. After the strategy is executed, the actual effect is continuously monitored, and the equipment operation parameters are adjusted in real time according to the deviation to form a dynamic closed-loop control. An infrared thermal imager with a resolution of 640x480 pixels is installed in the office area. Thermal imaging data is collected every 10 minutes to generate a temperature distribution map. The temperature distribution map is processed using the K-means clustering algorithm, and the number of clusters K=5 is set. The temperature value is clustered as a feature. The area that is 20% higher than the average ambient temperature is identified as an abnormal temperature area. If the average ambient temperature is 25℃, the area above 30℃ is marked as a concentrated area of ​​residual heat sources. Three temperature sensors, one flow sensor, and one pressure sensor are arranged within 1 meter around each identified residual heat source. The sensor collects data once a minute and transmits it to the data processing center through the ZigBee wireless network. A data integrity check mechanism is set up. If the received data packet is incomplete, it will automatically request retransmission. The sensor performs a self-check every hour to detect the power and communication status. The collected data is analyzed for time series. First, data preprocessing is performed, including outlier removal and missing value interpolation. Then trend analysis is performed, and the 7-day sliding average is calculated using the moving average method. Seasonal analysis is performed on the detrended data, and the autocorrelation function is used to identify periodic patterns. The ARIMA model is applied for time series prediction, setting the autoregressive term p=2, the difference term d=1, and the moving average term q=2. According to the prediction results and the current environmental conditions, the simplex method is used to solve the linear programming problem and calculate the optimal equipment operation parameters.If the predicted waste heat generation is greater than 5kW / h, the cooling device is started, and the cooling power is dynamically adjusted between 0-10kW; if it is less than 5kW / h, the heat absorption device is started, and the heating power is adjusted between 0-5kW. The strategy execution effect is evaluated every 15 minutes, and the root mean square error between the actual temperature and the target temperature is calculated. If the error exceeds 2°C, the parameters are re-optimized.

[0022] Statistics are collected on the waste heat distribution in each area, and the waste heat concentrated distribution areas are classified. Statistics are collected on the waste heat concentrated distribution areas to obtain statistical classification results of the waste heat concentrated distribution areas. The statistical classification results are used as the installation location strategy of the sensor network. According to the waste heat collection results of the sensor, a data collection strategy for the waste heat after the air conditioner is turned off in the office area is generated.

[0023] A multi-angle thermal imager is used to perform a temperature scan on the grid units, and a complete thermal map is synthesized through an image stitching algorithm to obtain a residual heat distribution thermal map; a cluster analysis is performed based on the residual heat distribution thermal map, and a hierarchical clustering algorithm is used to automatically determine the optimal number of clusters, and grid units with similar temperatures are classified into one category; if the number of grid units with similar temperatures is greater than a preset threshold, the temperature variance is less than a preset temperature threshold, the average temperature is higher than a preset difference from the ambient temperature, and the duration exceeds a preset time threshold, the grid units classified into one category are determined to be a highly concentrated area; the geometric center position of the highly concentrated area is obtained, and a temperature sensor is installed at the geometric center position, and the temperature sensor is integrated with the air-conditioning control system; the sampling frequency is dynamically adjusted according to the data change trend, and the collected data is compressed using a data compression algorithm, and the compressed data is transmitted to the central processor through a low-power wide area network.

[0024] Specifically, according to the pre-defined coordinate system of the office area plan, the entire area is divided into several grid units, each of which is 5 meters by 5 meters in size. A multi-angle thermal imager is used to scan the temperature of each grid unit, and a complete thermal map is synthesized through an image stitching algorithm to eliminate the occlusion caused by obstacles, obtain the average temperature value of each grid unit, and generate a residual heat distribution thermal map. Cluster analysis is performed on the residual heat distribution thermal map, and a hierarchical clustering algorithm is used to automatically determine the optimal number of clusters. Grid units with similar temperatures are classified into one category to obtain the preliminary classification results of the residual heat concentrated distribution area. The preliminary classification results are statistically analyzed to calculate the number of grid units, average temperature, temperature variance, and duration of each category. The degree of residual heat concentration is determined based on the comprehensive score, and the category with a number of grid units greater than 10, a temperature variance less than 2°C, an average temperature higher than the ambient temperature by more than 5°C, and a duration of more than 30 minutes is judged as a highly concentrated area. According to the spatial distribution of highly concentrated areas, a sensor network installation location strategy is formulated. A temperature sensor is installed at the geometric center of each highly concentrated area. It is integrated with the air conditioning control system to obtain the air conditioning operation status information in real time. The basic sampling interval is set to 5 minutes. When the air conditioning shutdown signal is detected or the temperature rise rate exceeds the preset threshold, the sampling interval is automatically adjusted to 1 minute. The residual heat change data is continuously recorded, and the sampling frequency is dynamically adjusted according to the data change trend. The data compression algorithm is used to reduce the storage space occupation, and the data is transmitted to the central processor through a low-power wide area network. In a 1,000-square-meter office area, the space is divided into 400 5-meter × 5-meter grid cells using a predefined coordinate system. A 360-degree rotating thermal imager with a resolution of 640 × 480 pixels is used to complete a full space scan every 15 seconds. Image stitching algorithms, such as SIFT feature matching and RANSAC algorithms, are used to synthesize multi-angle thermal imaging data into a complete thermal map to solve the occlusion caused by obstacles such as columns and partitions. The average temperature is calculated for each grid cell to generate a 40 × 40 temperature matrix. Ward's minimum variance method was used for hierarchical clustering. The cluster distance threshold was set to 2°C, and the optimal number of clusters was automatically determined. The clustering results were statistically analyzed, and the number of grid cells, average temperature, temperature variance, and duration of each category were calculated. The comprehensive scoring formula was set as follows: Score = number of grids × 0.3 + (average temperature - ambient temperature) × 0.4 + (10 - temperature variance) × 0.2 + duration × 0.1. Categories with scores greater than 15 were determined to be highly concentrated areas. A temperature sensor was installed at the geometric center of each highly concentrated area, with a measurement range of -20°C to 80°C and an accuracy of ±0.5°C. The sensor communicated with the air conditioning control system through the Modbus protocol to obtain the air conditioning switch status. The basic sampling interval was 5 minutes. When the air conditioning shutdown signal was detected or the temperature rise rate exceeded 0.5°C / minute, the sampling interval was adjusted to 1 minute.Adaptive sampling algorithm is used. When the temperature change rate is greater than 1℃ / 5 minutes, the sampling interval is halved; when it is less than 0.2℃ / 5 minutes, the sampling interval is doubled, with a minimum of 30 seconds and a maximum of 10 minutes. Differential coding compression algorithm is used to compress the original data to a ratio of 5:1. Through the LoRaWAN network, the data is transmitted to the central processor at a transmission rate of 5.5kbps in the 915MHz frequency band, achieving low power consumption and long-distance data transmission.

[0025] Step S103, retrieve the corresponding cold absorption or heat absorption operation parameters from the waste heat characteristic database, combine the dynamically adjusted cold absorption or heat absorption strategy, control the cold absorption or heat absorption equipment to adaptively adjust the operating state according to the parameter settings, and the operating state includes dynamically optimizing the heat pump speed and the expansion valve opening.

[0026] According to the acquired ambient temperature and humidity data, the waste heat feature record with the highest matching degree is retrieved from the waste heat feature database, and the waste heat feature database adopts a B-tree index structure. According to the waste heat feature record, the initial cooling or heat absorption operation parameters are obtained, and the initial cooling or heat absorption operation parameters include the heat pump speed range and the expansion valve opening range. The initial cooling or heat absorption operation parameters are optimized by a fuzzy logic controller, and the input variables of the fuzzy logic controller include temperature deviation and humidity deviation. The output frequency of the heat pump inverter is adjusted in real time by a PID controller, and the PID controller adopts the Ziegler-Nichols method to perform PID parameter self-tuning. The temperature, pressure, and flow data of the condenser and evaporator are obtained, and the temperature, pressure, and flow data are preprocessed by sliding average filtering. According to the preprocessed temperature, pressure, and flow data, the heat pump speed and expansion valve opening are fine-tuned in real time using an adaptive neural fuzzy inference system, and the adaptive neural fuzzy inference system continuously optimizes the inference rules through online learning.

[0027] Specifically, according to the current ambient temperature and humidity data, the waste heat feature record with the highest matching degree is retrieved from the waste heat feature database, the matching degree is calculated using cosine similarity, and the corresponding initial cooling or heat absorption operation parameters are extracted, including the heat pump speed range and the expansion valve opening range. The waste heat feature database adopts a B-tree index structure to optimize query efficiency. The initial operating parameters are integrated with the dynamically adjusted cooling or heat absorption strategy, and the fuzzy logic controller is used to optimize the parameters. The input variables include temperature deviation and humidity deviation. The rule base is set based on expert experience to obtain the heat pump speed target value and expansion valve opening target value that are suitable for the current environment. The output frequency of the heat pump inverter is adjusted in real time by the PID controller, and the PID parameter self-tuning is performed using the Ziegler-Nichols method to make the heat pump speed gradually approach the target value. At the same time, the stepper motor is used to accurately control the expansion valve opening so that it gradually reaches the target opening value. The operating status of the waste heat recovery system is continuously monitored. The temperature, pressure, flow and other parameters of the condenser and evaporator are collected every 5 seconds. The sliding average filter is used for data preprocessing. The adaptive neural fuzzy inference system (ANFIS) is used to fine-tune the heat pump speed and expansion valve opening in real time. The input of ANFIS is the preprocessed temperature, pressure and flow data, and the output is the speed adjustment and opening adjustment. The inference rules are continuously optimized through online learning to keep the waste heat recovery system in the optimal operating state. In an office building air conditioning system, the current ambient temperature is 26°C and the relative humidity is 60%. The waste heat feature database uses a B-tree index structure and stores 1,000 historical records. The waste heat recovery system uses the cosine similarity algorithm to calculate the matching degree, and the threshold is set to 0.95. The best matching record is retrieved, and the initial heat pump speed range is 30-60Hz and the expansion valve opening range is 40-80%. The fuzzy logic controller sets 5 input variables and 7 output variables, and the rule base contains 25 IF-THEN rules. After optimization, the target value of heat pump speed is 45Hz, and the target value of expansion valve opening is 60%. The PID controller uses the Ziegler-Nichols method to self-tune, and the parameters Kp=1.2, Ti=60s, and Td=15s are obtained. The waste heat recovery system collects operating data every 5 seconds, including condenser temperature, evaporator temperature, system pressure, and refrigerant flow. A 10-point sliding average filter is used for data preprocessing. The ANFIS system contains 5 input nodes, 25 rule nodes, and 2 output nodes. It uses the back propagation algorithm for online learning with a learning rate of 0.01. After 100 iterations, ANFIS outputs a fine-tuning of the heat pump speed by +2Hz and a fine-tuning of the expansion valve opening by -5%. The waste heat recovery system is continuously optimized based on feedback and reaches a stable state after 30 minutes. The indoor temperature is maintained at 24.5±0.5°C and the relative humidity is 50±3%.

[0028] Step S104, an energy metering device is deployed in the waste heat transmission pipeline network to monitor the cold and heat storage energy in the recovery process in real time, and analyze the energy loss value of the waste heat recovery. If it is found that the energy loss exceeds the preset threshold, the fault diagnosis program is triggered to determine the fault type and give a corresponding optimization control strategy.

[0029] Obtain the cold and hot fluid flow and temperature data of the key nodes of the waste heat transmission pipeline network, the cold and hot fluid flow and temperature data are collected by the ultrasonic flowmeter and PT100 temperature sensor installed at the key nodes; calculate the energy loss value between adjacent nodes based on the cold and hot fluid flow and temperature data, the energy loss value is calculated by the central controller based on the heat balance principle; determine whether the energy loss value exceeds the preset energy loss threshold, the preset energy loss threshold is determined based on the statistical analysis of historical data; if the energy loss value exceeds the preset energy loss threshold, trigger a fault diagnosis program, the fault diagnosis program uses the C4.5 decision tree algorithm to determine the fault type; according to the fault type determined by the fault diagnosis program, call the corresponding control strategy from the optimization control strategy library, the optimization control strategy library is constructed based on expert knowledge and machine learning algorithms; generate targeted control instructions, the control instructions are used to adjust the pump speed, change the valve opening or start the backup pipeline.

[0030] Specifically, ultrasonic flowmeters and PT100 temperature sensors are installed at key nodes of the waste heat transmission network to collect the flow and temperature data of cold and hot fluids in real time. The data is sampled every 10 seconds through the data collector, and the data is transmitted to the central controller using the ModBus protocol to calculate the instantaneous heat value of each node. At the same time, temperature sensors and liquid level gauges are installed on the energy storage equipment to monitor the changes in cold and hot storage energy. The central controller calculates the energy loss value between adjacent nodes based on the principle of heat balance. The loss calculation formula is Q=mc(T1-T2), where m is the mass flow rate of the fluid, c is the specific heat capacity, and T1 and T2 are the temperatures of adjacent nodes. Based on the statistical analysis of historical data, a preset energy loss threshold is set. If the energy loss of a certain section of the pipeline exceeds the threshold, the fault diagnosis program is triggered. The fault diagnosis program adopts the C4.5 decision tree algorithm, uses the information gain ratio for feature selection, combines multi-dimensional data such as temperature, pressure, and flow, judges the fault type, and divides the fault type into multiple categories such as pipeline leakage, insulation layer damage, and valve failure. According to the fault type obtained by diagnosis, the corresponding optimization control strategy library is called. The strategy library is built based on the combination of expert knowledge and machine learning algorithms and is continuously updated through online learning. Generate targeted control instructions, such as adjusting the speed of the pump, changing the valve opening or starting the backup pipeline, and apply the optimized control strategy to the waste heat recovery system through the actuator. In an industrial park waste heat recovery system, the total length of the pipeline network is 5 kilometers, and a total of 20 key nodes are set. Each node is equipped with an ultrasonic flow meter with an accuracy of ±0.5% and a PT100 temperature sensor with an accuracy of ±0.1°C. The data collector collects data through the ModBusRTU protocol at intervals of 10 seconds, with a transmission rate of 9600bps. The central controller uses an IntelCorei7 processor with an operating frequency of 3.6GHz and a memory of 32GB. In the heat balance calculation, the specific heat capacity of water is taken as 4.2kJ / (kg·K). Based on the historical data of the past three months, the waste heat recovery system uses the 3σ principle to set the energy loss threshold to 5%. The C4.5 decision tree algorithm uses information gain ratio for feature selection, sets the minimum number of leaf node samples to 50, and the maximum depth of the tree to 10. Fault types include pipeline leakage, insulation damage, valve failure, sensor failure, and pump failure. The optimization control strategy library initially contains 100 expert rules, uses the random forest algorithm for online learning, and updates the model every 24 hours. The waste heat recovery system detects that the energy loss of section A reaches 7.2%, triggering fault diagnosis. The decision tree analysis results show that the insulation layer is damaged, with a confidence level of 0.85. The control strategy automatically generates instructions: reduce the flow rate of this section of the pipeline by 20%, and start the backup pipeline at the same time. After receiving the instruction, the actuator adjusts the main pipeline pump frequency from 50Hz to 40Hz, and the backup pipeline valve opening from 0% to 60%. After 30 minutes of adjustment, the energy loss of section A dropped to 4.8%, and the waste heat recovery system resumed normal operation.

[0031] Step S105, obtain the real-time power generation data of the photovoltaic cell and the charge state data of the battery, combine the energy loss value of waste heat recovery, and build a photovoltaic power storage energy optimization scheduling model. The photovoltaic power storage energy optimization scheduling model includes factors such as photovoltaic power generation, battery capacity, and energy loss value of waste heat recovery. The energy allocation strategy is obtained by solving the photovoltaic power storage energy optimization scheduling model.

[0032] The real-time power generation data of the photovoltaic cells collected by the photovoltaic inverter is obtained, and the real-time power generation data includes output voltage, output current and output power; based on the real-time power generation data, a support vector regression algorithm is used to perform short-term prediction of the photovoltaic power generation within a preset time period in the future to obtain predicted power generation data; based on the predicted power generation data, the battery status data and the energy loss value, a multi-objective optimization function is constructed, and the multi-objective optimization function is F=w1f1+w2f2-w3f3, wherein f1 is the photovoltaic utilization rate, f2 is the battery utilization rate, f3 is the energy loss rate, and w1, w2, and w3 are weight coefficients; for the multi-objective optimization function, a genetic algorithm is used to solve and a Pareto optimal solution set is obtained; from the Pareto optimal solution set, a compromise optimal solution is selected by a hierarchical analysis method to determine an energy allocation strategy; according to the energy allocation strategy, a control instruction is generated, and the control instruction includes the photovoltaic power generation grid-connected power, the battery charging and discharging power and the waste heat recovery system operating parameters.

[0033] Specifically, the real-time power generation data of photovoltaic cells, including output voltage, current and power, are collected every 5 minutes through the photovoltaic inverter. At the same time, the state of charge data of the battery, including the current power and charge and discharge power, are obtained from the battery management system. These data are transmitted to the central controller through the communication gateway of the ModbusTCP protocol. The support vector regression algorithm is used to make a short-term forecast of the photovoltaic power generation in the next 4 hours, and the battery capacity change trend is analyzed based on historical data. The real-time energy loss value is obtained from the waste heat recovery system and used as the constraint condition of the optimization model. Combined with the photovoltaic power generation data, power generation prediction results, battery status data and its capacity change trend, a multi-objective optimization function F=w1f1+w2f2-w3*f3 is constructed, where f1 is the photovoltaic utilization rate, f2 is the battery utilization rate, f3 is the energy loss rate, w1, w2, and w3 are weight coefficients, respectively, to establish a photovoltaic power storage energy optimization scheduling model. The genetic algorithm is used to solve the photovoltaic power storage energy optimization scheduling model. The population size is set to 100, the crossover rate is 0.8, and the mutation rate is 0.1. The Pareto optimal solution set is obtained through 500 iterations. The hierarchical analysis method is used to select the optimal solution as the energy allocation strategy. According to the energy allocation strategy obtained by the optimization solution, specific control instructions are generated, including photovoltaic power grid-connected power, battery charging and discharging power, and operating parameters of the waste heat recovery system. These instructions are sent to the corresponding execution equipment through the controller to achieve energy optimization scheduling. In a smart industrial park, a photovoltaic power generation system with a total capacity of 1MW and a lithium battery energy storage system with a capacity of 500kWh are installed. The photovoltaic inverter collects data every 5 minutes to record the output voltage, current and power. The battery management system synchronously collects the battery charge state, including the current power and charging and discharging power. The data is transmitted to the central controller at a rate of 9600bps through the ModbusTCP protocol. The support vector regression algorithm uses historical data from the past 7 days and considers weather forecast information to predict the photovoltaic power generation in the next 4 hours. The average prediction error is controlled within ±10%. The analysis of the battery capacity change trend shows that the capacity decays by about 1% after every 100 complete charge and discharge cycles. The real-time energy loss value of the waste heat recovery system fluctuates between 5-15kW. In the multi-objective optimization function, the weight coefficients w1=0.5, w2=0.3, and w3=0.2. The genetic algorithm sets the population size to 100, the crossover rate to 0.8, and the mutation rate to 0.1. After 500 iterations, 20 Pareto optimal solutions are obtained. The hierarchical analysis method constructs a judgment matrix. After consistency testing, the solution with the highest comprehensive score is selected as the final energy allocation strategy. After the control command is generated, the photovoltaic grid-connected power is set to 850kW through the PLC, the battery charging power is 50kW, and the waste heat recovery system is adjusted to 70% load operation to achieve overall energy optimization scheduling. The waste heat recovery system recalculates every 15 minutes and dynamically adjusts the strategy.

[0034] Step S106, dynamically adjust the strategy for using cold and heat storage energy according to the energy allocation strategy obtained by solving, optimize the allocation ratio and timing of cold and heat energy, and realize the coordinated optimization control of photovoltaic power storage and waste heat recovery.

[0035] The real-time temperature, humidity, sunshine intensity data and the energy load forecast results based on the ARIMA model provided by the weather station are obtained; according to the data provided by the weather station and the energy load forecast results, a dynamic programming algorithm is used to calculate the usage sequence of cold and heat storage energy in the future preset time period, and the release of cold and heat energy in each time period is obtained; the dynamic programming algorithm takes energy balance as a constraint condition, takes the minimum total energy consumption as an objective function, the state variable is the energy level of the energy storage equipment, and the decision variable is the cold and heat energy released per hour; according to the cold and heat energy release, a fuzzy controller is used to dynamically adjust the operating parameters of the chiller and the hot water unit; the fuzzy controller The input variables of the controller are temperature deviation and load change rate, and the rule base includes IF-THEN rules; the real-time energy efficiency ratio of each device is obtained, and the distribution ratio of cold and hot energy is dynamically optimized according to the real-time energy efficiency ratio; the photovoltaic power generation and power load are predicted in real time through a three-layer feedforward neural network, the hidden layer of the three-layer feedforward neural network includes neurons, and the network is trained using the Levenberg-Marquardt algorithm; if the photovoltaic power generation is insufficient, it is discharged from the battery; if the battery power is lower than the preset threshold, the operating load of the waste heat recovery system is increased, and the power of non-critical energy-consuming equipment is reduced according to the preset priority order.

[0036] Specifically, according to the energy allocation strategy obtained, combined with the real-time temperature, humidity, sunshine intensity data provided by the meteorological station and the energy load forecast results based on the ARIMA model, the dynamic programming algorithm is used to calculate the time sequence of cold and heat storage energy use in the next 24 hours, and determine the amount of cold and heat energy released in each time period. The dynamic programming algorithm takes energy balance as a constraint condition, takes the minimum total energy consumption as the objective function, the state variable is the energy level of the energy storage equipment, and the decision variable is the cold and heat energy released per hour. Based on the calculated cold and heat energy release, the fuzzy controller is used to dynamically adjust the operating parameters of the chiller and hot water unit, including the outlet water temperature, flow rate and operating time. The input variables of the fuzzy controller are temperature deviation and load change rate, and the rule base contains 25 IF-THEN rules. At the same time, based on the real-time energy efficiency ratio of each device, the distribution ratio of cold and heat energy is dynamically optimized to improve the overall efficiency of the waste heat recovery system. The photovoltaic power generation and power load are predicted in real time through a three-layer feedforward neural network. The hidden layer contains 20 neurons, and the Levenberg-Marquardt algorithm is used to train the network. Combined with the battery charge state, the surplus or shortage of electric energy is calculated, and the operating power of the cold and heat storage equipment is adjusted accordingly to achieve electric and thermal synergistic optimization. A multi-level energy balance control strategy is set. When the photovoltaic power generation is insufficient, the battery is discharged first. If the battery power is less than 30%, the operating load of the waste heat recovery system is increased to 90%. At the same time, the power of non-critical energy-consuming equipment is reduced according to the preset priority order, and the load is reduced by 10%-30% to ensure the overall energy balance of the waste heat recovery system. In a certain intelligent building complex, a photovoltaic power generation system with a total capacity of 2MW, a lithium battery energy storage system with a capacity of 1MWh, and a waste heat recovery system with a capacity of 500kW are installed. The weather station provides real-time data every 5 minutes, including temperature, humidity, and sunshine intensity. The ARIMA model predicts the 24-hour energy load, and the average error is controlled within ±7%. The dynamic programming algorithm divides 24 hours into 96 15-minute periods, takes 0.5kWh as the energy unit, and constructs a state transfer matrix. The fuzzy controller uses a 5×5 rule matrix, and the input variables are discretized into 5 fuzzy sets. The input layer of the three-layer neural network contains 8 nodes, including time, temperature, humidity, sunshine intensity, etc., 20 nodes in the hidden layer, and 2 nodes in the output layer, including photovoltaic power generation and power load. The Levenberg-Marquardt algorithm was trained for 1,000 rounds, and the mean square error dropped below 0.01. The waste heat recovery system monitors the energy efficiency ratio of the chiller and hot water unit in real time. When the cold and hot load ratio is 3:2, the overall COP is increased by 8% by adjusting the cold water outlet temperature from 7°C to 9°C and the hot water from 55°C to 52°C. The battery discharge threshold is set to 30%. When the power drops to 28%, the waste heat recovery system load increases from 70% to 90%. At the same time, the lighting load, air conditioning load, and non-critical equipment load are reduced by 10%, 20%, and 30% respectively according to the preset order.The waste heat recovery system performs a global optimization calculation every 15 minutes and dynamically adjusts the operating parameters of each subsystem to maximize energy utilization efficiency.

[0037] Step S107, based on the acquired waste heat data, cooling or heat absorption strategy, optimization control strategy, and energy allocation strategy, a waste heat recovery system is constructed and continuously monitored, and the operation data of the waste heat recovery system is regularly analyzed to identify potential efficiency reduction or failure risks, and to adjust the operation parameters or perform maintenance in a timely manner.

[0038] The operating parameters of the waste heat recovery system are obtained, and the operating parameters include temperature, pressure and flow rate; a five-layer neural network model is constructed according to the operating parameters, and the input layer of the neural network model includes 20 nodes, the three hidden layers include 50, 30 and 20 neurons respectively, and the output layer includes 2 nodes; the operating data of the waste heat recovery system is collected through a real-time data acquisition system, and the operating data includes the inlet and outlet temperatures of the heat exchanger, the working fluid flow rate and the pressure parameters; the operating data is preprocessed, including: using a 5-minute sliding window average and a 3σ criterion to filter outliers; the preprocessed operating data is input into the five-layer neural network model to obtain the efficiency of the waste heat recovery system and the probability of waste heat recovery failure; it is determined whether the efficiency of the waste heat recovery system is lower than a preset threshold or whether the failure probability exceeds a preset threshold, if the efficiency of the waste heat recovery system is lower than the preset threshold or the failure probability exceeds the preset threshold, the long short-term memory network is triggered to perform fault diagnosis; according to the fault diagnosis result of the long short-term memory network, a fuzzy PID controller is used to adjust the operating parameters of the waste heat recovery system, and the operating parameters of the waste heat recovery system include valve opening and variable frequency pump speed.

[0039] Specifically, based on the acquired waste heat data, cold or heat absorption strategy, optimization control strategy and energy allocation strategy, a five-layer neural network model is constructed. The input layer contains 20 nodes, corresponding to parameters such as temperature, pressure, flow and strategy quantization values. The three hidden layers contain 50, 30 and 20 neurons respectively, using the ReLU activation function. The output layer has two nodes to predict the efficiency of the waste heat recovery system and the probability of waste heat recovery failure. The Adam optimizer and mean square error loss function are used for training with 50,000 historical data, and the learning rate is set to 0.001. Through the real-time data acquisition system, the operation data of the waste heat recovery system is collected once a minute, including parameters such as the inlet and outlet temperature of the heat exchanger, the working fluid flow rate, and the pressure. The outliers are filtered using a 5-minute sliding window average and the 3σ criterion. The pre-processed data is input into the neural network model to obtain the current waste heat recovery system efficiency and potential failure risk assessment results. The efficiency of the waste heat recovery system is defined as the ratio of output energy to input energy, and the failure risk is calculated based on the weighted score of indicators such as equipment temperature, pressure, and vibration. A multi-level threshold alarm mechanism is set up. When the efficiency of the waste heat recovery system is lower than 80% or the probability of failure exceeds 20%, the long short-term memory network is triggered to perform fault diagnosis, identify the specific fault type and location, and generate corresponding optimization adjustment suggestions or maintenance instructions. According to the fault diagnosis results and optimization suggestions, the fuzzy PID controller automatically adjusts the operating parameters of the waste heat recovery system, such as adjusting the valve opening, variable frequency pump speed, etc., with a control accuracy of ±2%. If the fault cannot be solved by parameter adjustment, a maintenance instruction is sent to the operation and maintenance management platform, and the fault information, historical data and diagnosis results are stored in the time series database. The performance report of the waste heat recovery system is automatically generated at 2 am every day, including 24-hour efficiency curve, energy consumption analysis and equipment status evaluation, and the performance changes of the waste heat recovery system in the next 7 days are predicted through trend analysis. In the waste heat recovery system of an industrial park, the input layer of the five-layer neural network model contains 20 nodes, corresponding to 10 temperature sensors, 5 pressure sensors, 3 flow meters and 2 policy quantization values. The three hidden layers contain 50, 30 and 20 neurons respectively, using the ReLU activation function. The two nodes in the output layer predict the efficiency of the waste heat recovery system and the probability of waste heat recovery failure respectively. The Adam optimizer is used with a learning rate of 0.001. 50 epochs are trained on the RTX3090 GPU, which takes about 2 hours. The real-time data acquisition system collects data once a minute and uses a 5-minute sliding window average and a 3σ criterion to filter outliers. When the efficiency of the waste heat recovery system is less than 80% or the probability of failure exceeds 20%, the LSTM network is triggered for fault diagnosis. The LSTM contains 64 hidden units and the input is the time series data of the past 24 hours. The fuzzy rule base of the fuzzy PID controller contains 25 rules. The input variables are deviation and deviation change rate, and the output is the adjustment amplitude of the control amount. The time series database uses InfluxDB, which writes 200 data points per second.The daily report includes 24-hour efficiency curve, energy consumption analysis and equipment status assessment. The trend analysis uses the ARIMA model to predict the efficiency of the waste heat recovery system in the next 7 days, with an average error of less than ±5%.

[0040] It will be apparent to those skilled in the art that the present application is not limited to the details of the exemplary embodiments described above, and that the present application can be implemented in other specific forms without departing from the spirit or essential features of the present application. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the present application is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present application. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. An automatic waste heat recovery method, characterized in that: include: Obtain waste heat pressure parameters, component medium and flow characteristic data, classify waste heat data, obtain characteristic vectors of different types of waste heat, including temperature, pressure, flow, and heat capacity parameters, and store characteristic vectors of different types of preheating in a waste heat characteristic database; Obtain the waste heat distribution inside and outside the office environment, identify the concentrated areas of each waste heat source, collect the generation data of each waste heat source in real time, analyze and predict the randomness and intermittent laws of waste heat generation and the waste heat concentrated areas, and dynamically adjust the cooling or heat absorption strategy; Retrieving corresponding cold absorption or heat absorption operation parameters from the waste heat characteristic database, and controlling the cold absorption or heat absorption equipment to adaptively adjust the operation state according to the parameter settings; Real-time monitoring of the cold and heat storage energy during the recovery process is carried out, and the energy loss value of the remaining heat recovery is analyzed. If the energy loss exceeds the preset threshold, the fault diagnosis program is triggered to determine the fault type and give the corresponding optimization control strategy; Construct a photovoltaic power storage energy optimization scheduling model, solve the energy allocation strategy, dynamically adjust the use strategy of cold and heat storage energy, optimize the distribution ratio and timing of cold and heat energy, and realize the coordinated optimization control of photovoltaic power storage and waste heat recovery.

2. The method according to claim 1, characterized in that The obtaining of waste heat pressure parameters, component medium and flow characteristic data includes: Obtain the temperature parameters, pressure characteristics, flow data and heat capacity properties of the waste heat recovery system, perform a value range check on the parameter data, and mark it as abnormal data if the parameter value exceeds the preset threshold range; Performing linear interpolation processing or elimination on the abnormal data, and normalizing the normal data to obtain standardized parameter values; The waste heat medium is subjected to composition analysis by using a gas chromatograph to identify the main chemical components and their contents, and the composition data is combined with the standardized parameter values.

3. The method according to claim 1, characterized in that The classification of waste heat data includes: The K-means clustering algorithm is used to classify the merged multidimensional data to obtain the characteristic clusters of different types of residual heat, and the centroid coordinates of each characteristic cluster are calculated as the characteristic vector of the residual heat; A waste heat characteristic database table is established, which includes waste heat type, temperature vector, pressure vector, flow vector, heat capacity vector and component vector, and characteristic vectors of different types of waste heat are inserted into the waste heat characteristic database table for storage.

4. The method according to claim 1, characterized in that The obtaining of the waste heat distribution inside and outside the office environment includes: Obtaining an ambient temperature distribution map, wherein the ambient temperature distribution map is obtained by acquiring real-time data of multiple temperature sensors around the exhaust port of the air compressor outside the office environment and the air conditioner condenser inside the office environment, and estimating the waste heat distribution inside and outside the office environment using the Kriging interpolation algorithm according to the exhaust temperature and exhaust flow rate of the air compressor and the exhaust temperature and exhaust flow rate parameters of the air conditioner condenser; Executing a K-means clustering algorithm according to the ambient temperature distribution map to obtain an abnormal temperature area as a waste heat source concentration area; A sensor network is arranged around the waste heat source concentrated area, and the sensor network includes a temperature sensor, a flow sensor and a pressure sensor.

5. The method according to claim 1, characterized in that The analysis predicts the randomness and intermittent patterns of waste heat generation and waste heat concentration areas, including: Collecting waste heat source data acquired by a sensor network, wherein the waste heat source data includes temperature data, flow data and pressure data; Performing time series analysis on the waste heat source data, wherein the time series analysis includes data preprocessing, trend analysis and seasonal analysis; The waste heat source data is processed using an autoregressive moving average model to obtain a characteristic curve and a prediction model for each waste heat source.

6. The method according to claim 1, characterized in that The dynamic adjustment of the cold absorption strategy or the heat absorption strategy includes: Calculating an optimal cold or heat absorption strategy according to the characteristic curve and the prediction model, wherein the optimal cold or heat absorption strategy is obtained by a linear programming method; If the waste heat generated by the prediction model is greater than a preset threshold, the cooling device is started and the cooling power is adjusted; If the waste heat generated by the prediction model is less than a preset threshold, the heat absorbing device is started and the heating power is adjusted.

7. The method according to claim 1, characterized in that The step of retrieving corresponding cold absorption or heat absorption operation parameters from the waste heat characteristic database includes: According to the acquired ambient temperature and humidity data, the residual heat feature record with the highest matching degree is retrieved from the residual heat feature database, wherein the residual heat feature database adopts a B-tree index structure; According to the waste heat characteristic record, initial cooling or heat absorption operation parameters are obtained, wherein the initial cooling or heat absorption operation parameters include a heat pump speed range and an expansion valve opening range; A fuzzy logic controller is used to optimize the initial cooling or heating operation parameters, and the input variables of the fuzzy logic controller include temperature deviation and humidity deviation.

8. The method according to claim 1, characterized in that The control of the cold or heat absorbing device to adaptively adjust the operating state according to the parameter setting includes: The output frequency of the heat pump inverter is adjusted in real time by a PID controller, wherein the PID controller uses the Ziegler-Nichols method to perform PID parameter self-tuning; Acquire the temperature, pressure and flow data of the condenser and the evaporator, wherein the temperature, pressure and flow data are preprocessed by using a sliding average filter; According to the pre-processed temperature, pressure and flow data, an adaptive neural fuzzy inference system is used to perform real-time fine-tuning on the heat pump speed and expansion valve opening. The adaptive neural fuzzy inference system continuously optimizes the inference rules through online learning.

Images