Multi-parameter cold source intelligent control system and method based on big data

Abstract

The invention provides a multi-parameter cold source intelligent control system and method based on big data, and the system comprises a data collection and control module which is used for collecting related parameters of a cold source system; the comprehensive energy consumption standardization module is used for converting the energy consumption parameters into unified standardized energy consumption indexes; the comfort degree algorithm construction module is used for generating a dynamic comfort degree index; the association rule mining module is used for mining a multi-dimensional association rule from historical data; the intelligent decision module is used for generating an optimal control strategy based on the index, the index and the association rule; according to the embodiment of the invention, based on a multi-parameter fusion cold source intelligent control technology of deep reinforcement learning and an improved FP-Growth algorithm, through deep mining and intelligent decision making of multi-source data, the optimal control strategy is converted into equipment regulation and control parameters, and the equipment regulation and control parameters are obtained. And high efficiency and energy conservation of the cold source system and accurate regulation and control of indoor environment comfort are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent control, in particular to a multi-parameter cold source intelligent control system and method based on big data. BACKGROUND

[0002] With the continuous rise of global building energy consumption proportion, the dynamic optimization control of central air conditioning system as the core hub of building energy system has become the key breakthrough to realize building energy saving. The PID control technology used in traditional boilers and refrigeration units has obvious limitations: on the one hand, its response speed lags behind the actual load change, resulting in energy waste; on the other hand, relying on manual experience to adjust parameters is not only inefficient, but also difficult to cope with the complex working conditions of multi-variable coupling and non-linear change in modern buildings. It is worth noting that about 30% of the energy consumption in the air conditioning system operation process is consumed in unnecessary working conditions, such as continuous operation during unattended period, excessive refrigeration / heating, etc. However, the traditional energy consumption management method can only monitor surface data, and lacks the intelligent mining ability of the deep correlation rules between equipment operating parameters, environmental factors and energy consumption, which seriously restricts the full release of energy saving potential.

[0003] The publication number CN118132968A proposes a tin smelting energy consumption data acquisition system and analysis method based on Apriori algorithm. Although this application establishes the correlation rules between energy consumption and environmental factors through multi-element data acquisition equipment such as electric quantity, temperature, pressure, natural gas and water consumption, and extracts regression coefficients to construct a multiple regression model, it can only realize the quantitative analysis of energy consumption rules, and lacks real-time control ability of the equipment, and cannot realize the energy saving goal through dynamic optimization of equipment operating parameters.

[0004] The publication number CN120295180A proposes a central air conditioning cold and heat source energy saving intelligent control system. Through the built-in intelligent control system proposed in this scheme, the intelligent control start can be effectively combined with the user's control demand time point, control temperature and environmental temperature in the control process of central air conditioning, but the system cannot automatically learn from the user's energy use habits, and cannot consider other related factors that affect the efficiency of central air conditioning, such as working time, personnel activity, etc. Therefore, there is an urgent need for a multi-parameter fusion cold source intelligent control technology to realize efficient energy saving of the cold source system and precise regulation of indoor environmental comfort. SUMMARY

[0005] The present application provides a multi-parameter cold source intelligent control system and method based on big data, which realizes efficient energy saving of the cold source system and precise regulation of indoor environmental comfort, and solves the problem of lack of intelligent mining ability of deep correlation rules between equipment operating parameters, environmental factors and energy consumption mentioned in the above background technology.

[0006] In a first aspect, the embodiments of the present specification describe a multi-parameter cold source intelligent control system based on big data, comprising:

[0007] a data acquisition and control module for acquiring cold source system energy consumption parameters, environmental parameters, personnel activity parameters, equipment operating state parameters, and executing control instructions;

[0008] a comprehensive energy consumption standardization module connected with the data acquisition and control module, for receiving energy consumption parameters, and converting the energy consumption parameters into unified standardized energy consumption indicators, and outputting energy efficiency evaluation indicators;

[0009] a comfort level algorithm construction module connected with the data acquisition and control module, for receiving the environmental parameters and personnel activity parameters, and generating a dynamic comfort level index;

[0010] an association rule mining module connected with the data acquisition and control module, the comprehensive energy consumption standardization module, and the comfort level algorithm construction module, for mining multi-dimensional association rules between the environmental parameters, the standardized energy consumption indicators, and the comfort level index from historical data;

[0011] an intelligent decision-making module connected with the comprehensive energy consumption standardization module, the comfort level algorithm construction module, and the association rule mining module, for generating an optimal control strategy based on the standardized energy consumption indicators, the comfort level index, and the association rules;

[0012] a control instruction generation module connected with the intelligent decision-making module and the data acquisition and control module, for converting the optimal control strategy into equipment control parameters, and sending it to the data acquisition and control module to drive the cold source system to execute the control instructions. By adopting the above technical solution, a multi-modal data acquisition network is constructed, and the cold source running state, environmental parameters, and load demand are monitored comprehensively and with high precision. The comfort level algorithm construction module accurately quantifies human thermal perception according to international standards, optimizes energy efficiency and comfort simultaneously, and significantly improves system decision-making accuracy and response speed. The intelligent decision-making module mines deep energy consumption rules, outputs association rules, supports adaptive dynamic optimization of control strategies, and realizes rapid and accurate regulation of equipment parameters. The control instruction generation module ensures efficient execution of strategies, realizes rapid cooling and steady-state maintenance through intelligent adjustment, and achieves the goal of energy saving and efficiency improvement.

[0013] Further, the data acquisition and control module comprises:

[0014] an energy metering unit for monitoring the consumption of electric energy, natural gas, and water resources;

[0015] a thermal state monitoring unit for monitoring the three-dimensional temperature field of the evaporator and the condenser and the phase change state of the refrigerant;

[0016] An environment perception unit is configured to monitor indoor light intensity, indoor carbon dioxide concentration, and personnel density and distribution. By adopting the above technical solution, the multi-dimensional unit is integrated, comprehensive and refined monitoring of energy consumption, thermal state and environmental parameters is achieved, rich and accurate data basis is provided for subsequent intelligent decision-making, and the system perception ability and control accuracy are improved.

[0017] Further, the comprehensive energy consumption standardization module is configured to convert electricity, natural gas, water and other heterogeneous energy into standard coal equivalent according to their heat value equivalent by constructing a multi-energy unified evaluation system, the multi-energy unified evaluation system is realized based on a dynamic conversion model, and the dynamic conversion model is: wherein E represents total standard coal equivalent, a_i represents the conversion coefficient of the i-th energy, and Q_i represents the sub-energy consumption of the i-th energy. By constructing a multi-energy unified evaluation system and adopting a dynamic conversion model, the standardization and quantitative evaluation of electricity, natural gas, water and other heterogeneous energy consumption are realized, and scientific index basis is provided for system energy efficiency accurate analysis and optimization.

[0018] Further, the comprehensive energy consumption standardization module outputs the energy efficiency evaluation index in real time as: wherein E_t represents the energy consumption standard coal equivalent at the current time, and E_{min} represents the historical minimum energy consumption standard coal equivalent. By calculating the energy efficiency evaluation index in real time, the system can dynamically reflect the gap between the current energy consumption level and the historical optimal level, and provide immediate and quantitative reference basis for energy efficiency monitoring, fault diagnosis and optimization adjustment.

[0019] Further, the comfort degree algorithm construction module calculates the dynamic comfort degree index based on the PMV-PPD model, and constructs a comfort degree reward item based on the dynamic comfort degree index, gives a positive reward when the predicted percentage of dissatisfied personnel PPD value is lower than a threshold value, and gives a negative reward when the predicted percentage of dissatisfied personnel PPD value is higher than a threshold value. By adopting the PMV-PPD model and constructing the comfort degree reward mechanism, the personnel thermal comfort is quantified and included in the optimization target, and the accurate perception and positive guidance of comfort degree are realized.

[0020] Further, the association rule mining module adopts an improved FP-Growth algorithm to mine a high-frequency association rule set and extract non-obvious energy consumption association rules. By adopting the above technical solution, the system can identify deep running rules and provide interpretable and reliable data support for intelligent decision-making.

[0021] Further, the intelligent decision module adopts a deep reinforcement learning algorithm to output compressor frequency and electronic expansion valve opening action instructions with energy consumption and PMV comfort index as reward functions. By adopting the deep reinforcement learning algorithm, the system can adaptively generate key control instructions such as compressor frequency and electronic expansion valve opening to realize precise control under complex working conditions, with energy consumption and comfort as common optimization targets.

[0022] Further, the reward function R is composed of a comfort reward term R_comfort and an energy consumption reward term R_energy, and the formula is: where ω_1 and ω_2 are weight coefficients. By designing the weighted reward function, the system can flexibly balance the two goals of energy saving and comfort, and adjust the weight coefficients according to different scenes to realize the configurability and multi-objective collaborative optimization of the control strategy.

[0023] In a second aspect, the embodiments of the present specification provide a multi-parameter cold source intelligent control method based on big data, including the steps of:

[0024] Collecting cold source system operation energy consumption parameters, environmental parameters, personnel activity parameters, and equipment operation state parameters, and executing control instructions;

[0025] Converting different types of energy consumption parameters into standardized energy consumption indicators;

[0026] Generating a dynamic comfort index based on the environmental parameters and personnel activity parameters;

[0027] Based on historical operation data, mining multi-dimensional association rules between operation environmental parameters, standardized energy consumption indicators, and comfort indexes;

[0028] Taking the standardized energy consumption indicators and the comfort indexes as optimization targets, and combining the association rules, generating an optimal control strategy through an intelligent learning algorithm;

[0029] Converting the optimal control strategy into specific equipment control parameters and executing them.

[0030] Further, the intelligent learning algorithm is a deep reinforcement learning algorithm; when the mined association rules trigger a specific emergency working condition, an emergency control strategy is started; the deep reinforcement learning algorithm adopts a priority experience replay mechanism to give higher weights to training samples corresponding to emergency working conditions.

[0031] The technical solutions provided by some embodiments of the present specification have at least the following beneficial effects:

[0032] In the embodiments of the present specification, the provided big data-based multi-parameter cold source intelligent control system and method realize comprehensive and high-precision perception of the running state of the cold source system, environmental conditions and load changes by introducing a multi-modal data acquisition network, the comfort degree algorithm construction module realizes double-target balanced optimization of energy efficiency and comfort degree based on the quantification of human thermal comfort degree according to international standards, and the accuracy and responsiveness of the decision are improved; the deep reinforcement learning decision module combines the improved FP-Growth algorithm to mine non-obvious energy consumption association rules and output the association rules, realizes adaptive dynamic optimization of the cold source system control strategy, has excellent performance in processing dynamic load changes and complex coupling relationships, can quickly respond and accurately adjust the equipment running parameters, realizes precise control of the cold source system, effectively reduces energy consumption, and improves energy utilization efficiency. At the same time, the control instruction generation module ensures the accurate execution of the deep reinforcement learning strategy, realizes the smooth switching of rapid cooling and constant temperature maintenance through dynamically adjusting the evaporation temperature set value and the like.

[0033] Other features and advantages of the embodiments of the present specification will be further disclosed in the following specific embodiments and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present specification, the drawings required to be used in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present specification, and other drawings can also be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0035] Figure 1 A big data-based multi-parameter cold source intelligent control system schematic diagram provided for the embodiments of the present specification.

[0036] Figure 2 A deep learning process schematic diagram in the intelligent decision module provided for the embodiments of the present specification.

[0037] Figure 3 A big data-based multi-parameter cold source intelligent control method flowchart provided for the embodiments of the present specification.

[0038] Figure 4 A dynamic comfort degree index generation flowchart provided for the embodiments of the present specification. DETAILED DESCRIPTION

[0039] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments of the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative work fall within the scope of the present application. The terms "first", "second", "third", etc. in the description, claims and above drawings are used to distinguish different objects, but not to describe a particular order. In addition, the terms "include" and "have" and any variations thereof are intended to cover the inclusions without exclusivity. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but optionally includes other steps or units not listed or optionally includes other steps or units inherent to the process, method, product or device.

[0040] In the following description, the orientation or position relationship such as the terms "inner", "outer", "upper", "lower", "left", "right", etc. is only for the convenience of describing the embodiments and simplifying the description, and does not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present specification.

[0041] Please refer to Figure 1 The present specification provides a multi-parameter cold source intelligent control system based on big data, which includes six core functional modules. These modules are tightly coupled and work together to form a closed-loop intelligent control system. The six core functional modules are: data acquisition and control module 100, comprehensive energy consumption standardization module 200, comfort algorithm construction module 300, association rule mining module 400, intelligent decision module 500 and control instruction generation module 600. The entire system responds to environmental and load changes in real time through data flow and iterative optimization between modules, and dynamically adjusts the operating state of the cold source equipment.

[0042] a data acquisition and control module 100, which is configured with a multi-modal sensor array and an actuator, for acquiring cold source system energy consumption parameters, environmental parameters, personnel activity parameters, equipment operating state parameters, and receiving and executing control instructions from the control instruction generation module; the data acquisition and control module is constructed with a multi-modal perception network oriented to cold source system optimization, which realizes the acquisition of characteristic parameters required by the algorithm through an energy metering unit, a thermal state monitoring unit and an environmental perception unit. Specifically, it includes but is not limited to the acquisition of temperature (indoor / outdoor temperature difference gradient), humidity (relative humidity and dew point temperature), personnel density (infrared perception and access control data fusion), solar intensity (ultraviolet index and azimuth), weather type (precipitation probability and wind speed). Further, the energy metering unit is configured with an intelligent electric meter with a precision of 0.5 level, a G1.6 level gas flow meter and an ultrasonic water meter with a precision of ±1%, for realizing the monitoring of device-level electric energy, natural gas and water resource consumption; the thermal state monitoring unit is configured with a PT100 platinum resistance array with A-level precision and a digital pressure sensor with a range of 0-25MPa, for capturing the three-dimensional temperature field of the evaporator and the condenser and tracking the phase change process of the refrigerant in real time; the environmental perception unit is integrated with an illuminance sensor with a range of 0-200klux, a CO2 sensor based on the NDIR principle and a binocular vision system with a precision of ±3%, for measuring indoor light intensity, monitoring indoor carbon dioxide concentration and establishing indoor personnel density and spatial distribution model. Specifically, the data acquisition is carried out through the industrial Internet of Things architecture, using the MODBUS-TCP protocol for millisecond-level acquisition, the preprocessing includes accessing the real-time weather data of the meteorological bureau through the HTTP RESTful API, and aligning the time stamp and filling the missing values of all data, the structured time series data set contains 15-dimensional state space features, which are pushed to the cloud time series database at a sampling frequency of 200Hz through the MQTT protocol, meeting the needs of deep reinforcement learning for high-dimensional and high-timeliness training data.

[0043] A comprehensive energy consumption standardization module 200 connected with the data acquisition and control module, for receiving energy consumption parameters and converting energy consumption parameters into unified standardized energy consumption indicators and outputting energy efficiency evaluation indicators; specifically, according to GB / T2589-2020 standard, the multi-source energy consumption such as electricity, natural gas and water is unified converted into kilogram standard coal equivalent, so as to establish a comparable multi-energy unified evaluation system; the multi-energy unified evaluation system is realized based on a dynamic conversion model, and the dynamic conversion model is: wherein E represents the total standard coal equivalent, a_i represents the conversion coefficient of the i-th energy, and Q_i represents the sub-energy consumption of the i-th energy. Through energy type normalization processing, the original 7-dimensional energy consumption parameter is compressed into a 1-dimensional standard coal index, reducing the state space dimension of deep reinforcement learning by 85.7%; as the core calculation unit of the reward function, the above technology can output the energy efficiency evaluation index in real time , which provides the gradient update direction for the policy network. Wherein E_t represents the energy consumption standard coal equivalent at the current time, E_{min} represents the historical minimum energy consumption standard coal equivalent, and the energy efficiency evaluation index provides the gradient update direction for the policy network. Further, the comprehensive energy consumption standardization module automatically corrects the conversion coefficient (such as the seasonal correction factor β = 0.92-1.15) combined with the association rules mined by the FP-Growth algorithm, so that the standard coal calculation error is controlled within ±2.3%. According to the actual measurement, this module improves the convergence speed of reinforcement learning by 40%, and the total standard coal calculation can be used as the energy efficiency benchmark, supporting the ISO50001 energy management system certification. The current step not only serves the purpose of dimension reduction, but also provides quantitative energy consumption reward input for deep reinforcement learning.

[0044] The comfort algorithm construction module 300 is connected with the data acquisition and control module, and is used for receiving the environmental parameters and personnel activity parameters, and constructing a PMV-PPD model to quantify the human thermal comfort and generate a dynamic comfort index combined with indoor and outdoor environmental parameters; the comfort algorithm construction module is provided with a reward function in the form of multi-objective weighted aggregation in the deep reinforcement learning framework, the reward function is designed in the form of multi-objective weighted aggregation, and the formula of the reward function R is: where ω_1 and ω_2 are weight coefficients, the comfort reward R_comfort is constructed according to the ISO7730:2023 international standard to build a dynamic PMV-PPD model to calculate the predicted mean vote (PMV) and the predicted percentage of dissatisfied (PPD). A positive reward is given when the PPD value is below 10%, and a negative punishment is given according to ΔPPD / 15 when the threshold is exceeded. The energy consumption reward R_energy is unified and quantified by a standard coal equivalent conversion module, which has a built-in conversion coefficient library of “General Energy Consumption Calculation” (GB / T 2589-2020). The module converts heterogeneous energy data such as electricity (0.1229 kgce / kWh) and natural gas (1.2143 kgce / m³) into comparable standard coal values in real time, and uses a sliding window algorithm to calculate the energy consumption change rate in the past 24 hours as the basis for rewards and punishments. The FP-Growth algorithm mines the device parameter association rules (such as the frequent item set support of cooling water flow and chilled water outlet temperature is 0.32) from historical operation data. The generated feature vector will be used as the dimension reduction input of the deep reinforcement learning state space, and finally form an adaptive control strategy that meets the LEED v4.1 certification requirements. The current step not only serves the purpose of dimension reduction, but also provides a quantitative comfort reward input for deep reinforcement learning.

[0045] The association rule mining module 400 is connected with the data acquisition and control module, the comprehensive energy consumption standardization module, and the comfort algorithm construction module. It analyzes historical data through an improved frequent pattern mining algorithm (single scan + vertical data format) to mine multi-dimensional association rules between the environmental parameters, standardized energy consumption indicators, and comfort indexes from the historical data. The association rule mining module introduces a dynamic minimum support threshold function to adaptively adjust the threshold according to the parameter dimension (such as setting the personnel density weight coefficient to 0.3 and the temperature weight to 0.25). It uses matrix compression storage technology to optimize the FP-tree construction process, reducing memory usage by about 40%. The high-frequency association rule set output by the algorithm, such as the confidence of the refrigeration load surge under the condition of sunlight > 800 W / m² and humidity > 70% is 92%, and the confidence of the refrigeration demand increase under the condition of temperature ≥ 28℃ and humidity > 65% is 90%.

[0046] The intelligent decision module 500 is connected with the comprehensive energy consumption standardization module, the comfort degree algorithm construction module and the association rule mining module. The intelligent decision module adopts a deep reinforcement learning algorithm, takes energy consumption (standard coal) and PMV comfort degree index (ISO7730:2023) as a reward function, and outputs compressor frequency, electronic expansion valve opening degree and other action instructions. The intelligent decision module takes the output result of the cold and heat demand prediction model as input, combines real-time temperature, humidity, cleanliness and air flow speed and other parameters in the room, and takes the comfort degree evaluated according to the ISO7730:2023-04 international thermal comfort environment standard as a reward function. The reinforcement learning agent continuously learns and optimizes the decision strategy through interaction with the environment. When control is needed, the intelligent output corresponding control actions, including adjusting the running frequency of the compressor, the opening degree of the electronic expansion valve, controlling the running state of the heating and ventilation system, and adjusting the temperature, humidity, cleanliness and air flow speed and other parameters, to achieve the goal of rapid cooling or constant temperature maintenance.

[0047] Please refer to Figure 2 The deep reinforcement learning steps are as follows:

[0048] The reward function is designed according to the energy consumption reward and the comfort degree reward. It needs to be specially pointed out that the negative logarithm of the standard coal consumption (kgce / h) is calculated in real time as the energy consumption reward item, and the comfort degree reward item is based on the ISO7730:2023 standard to construct a PMV index dynamic reward function. When PMV ∈ [-0.5, 0.5], a positive reward is given;

[0049] When the frequent pattern recognition triggers the [outdoor temperature > 35℃ ∩ human flow surge 40%] rule, the agent immediately starts the emergency control strategy, wherein the emergency control strategy is:

[0050] The compressor frequency is increased from 35Hz to 42Hz (slope 0.23Hz / s) within 30 seconds;

[0051] The electronic expansion valve opening degree is adjusted to 65%±2% at the same time;

[0052] The COP=4.8±0.3 and PMV=0.2±0.1 are maintained by PID compensation;

[0053] For state-action space modeling, specifically,

[0054] The load data and real-time sensor data (temperature / humidity / cleanliness / air flow speed) fused by the cold and heat demand prediction model are taken as the input layer;

[0055] The continuous control instructions of the compressor frequency (20-50Hz) and the electronic expansion valve opening degree (10%-90%) are taken as the output layer;

[0056] Online learning optimization is performed, a priority experience replay mechanism is adopted, 3 times sampling weight is given to the emergency working condition sample triggered by FP-Growth, and the strategy convergence is accelerated.

[0057] The intelligent decision module takes the high-frequency association rule set output by the association mining module as prior knowledge or state space dimension reduction input, and gives higher learning weight to the working condition sample triggering a specific association rule, takes the energy efficiency evaluation index and the dynamic comfort index as the core components of the reward function, constructs a double-objective optimization model based on deep reinforcement learning, and generates an optimal control strategy considering energy efficiency (lowest energy consumption) and comfort (PMV≈0).

[0058] The control instruction generation module 600 is connected with the intelligent decision module and the data acquisition and control module, and is used for converting the optimal control strategy output by the deep reinforcement learning decision module into specific device control parameters (such as compressor frequency and water pump flow), and sending the specific device control parameters to the data acquisition and control module to drive the cold source system to execute the control instructions. The system dynamically iterates at a period of 15 minutes, realizes adaptive control of the cold source system through real-time rule updating of FP-Growth and online strategy optimization of deep reinforcement learning.

[0059] Please refer to Figure 3 The embodiment of the present specification also provides a multi-parameter cold source intelligent control method based on big data, comprising the following steps:

[0060] Step S201: data acquisition and preprocessing, real-time acquisition of cold source system running environment parameters, device running state parameters, energy consumption parameters and personnel activity parameters through a multi-modal sensor array, and data cleaning, calibration and synchronization to form a structured time series data set.

[0061] Step S202: converting different types of energy consumption parameters into standardized energy consumption indicators. According to the national standard GB / T2589-2020, the energy consumption parameters collected in step S201 are dynamically converted, and heterogeneous energy such as electricity, natural gas and water is converted into standard coal equivalent, and the energy efficiency evaluation index is calculated.

[0062] Step S203: comfort index construction, according to the international standard ISO7730:2023, based on the environmental parameters and personnel activity parameters collected in step S201, the dynamic comfort index is calculated and generated; please refer to Figure 4 The generation of the dynamic comfort index includes:

[0063] The dynamic comfort index is calculated by a PMV-PPD model, and a comfort reward term is constructed based on the dynamic comfort index, a positive reward is given when the predicted percentage of dissatisfied (PPD) value is below a threshold, and a negative reward is given when the predicted PPD value is above the threshold. The PMV-PPD model calculates a predicted mean vote (PMV) value by the following formula.

[0064] ,

[0065] wherein M is the human energy metabolic rate, W is the human external work amount, H is the human convective heat dissipation amount, E is the human evaporative heat dissipation amount, C is the human convective heat exchange amount with the surrounding environment, and R is the human radiative heat exchange amount with the surrounding environment; the PMV value is further converted into a predicted percentage of dissatisfied (PPD) value by the following formula:

[0066] ,

[0067] The PMV value and the PPD value are calculated based on real-time collected environmental parameter data and user personalized data, and the dynamic comfort index is used to construct a comfort reward term R_comfort of deep reinforcement learning, and the reward mechanism is that a positive reward is given when the PPD value is below 10%, and a negative punishment is given according to ΔPPD / 15 when the PPD value exceeds the threshold.

[0068] Step S204: association rule mining, performing improved FP-Growth algorithm association analysis on the historical operation data collected in step S201, mining multi-dimensional association rules between environmental parameters, energy consumption indicators and comfort indexes, and outputting a high-frequency association rule set.

[0069] Step S205: intelligent decision-making, taking the energy efficiency evaluation indicators calculated in step S202 and the dynamic comfort index generated in step S203 as the core components of the reward function, and taking the high-frequency association rule set output in step S204 as prior knowledge or state space dimension reduction input, constructing an intelligent learning algorithm, wherein the intelligent learning algorithm is a deep reinforcement learning model, which learns online and generates an optimal control strategy considering energy efficiency and comfort through interaction with the cold source system environment; the reward function of the deep reinforcement learning algorithm takes the energy consumption target reward term as the negative logarithm of the standard coal consumption (kgce / h). At the same time, when the FP-Growth association rule mining result of step S204 triggers a specific emergency working condition rule, the agent of the deep reinforcement learning algorithm will immediately start an emergency control strategy, and the deep reinforcement learning model adopts a priority experience replay mechanism, and gives a 3-fold sampling weight to the emergency working condition sample triggered by the FP-Growth, so as to accelerate the strategy convergence.

[0070] Step S206: control instruction generation and execution, the optimal control strategy generated in step S205 is converted into specific device control parameters, and is sent to the executor of the data acquisition and control module to adjust the compressor frequency, electronic expansion valve opening, water pump flow and other cold source device parameters.

[0071] The above is only an example and explanation of the concept of the present application. Those skilled in the art can make various modifications or supplements to the described specific embodiments or use similar ways to replace them, as long as they do not deviate from the concept of the present application or exceed the scope defined by the present claims.

Claims

1. A multi-parameter intelligent control system for cold sources based on big data, characterized in that, include: The data acquisition and control module is used to collect energy consumption parameters, environmental parameters, personnel activity parameters, and equipment operating status parameters of the cold source system, and execute control commands. The integrated energy consumption standardization module, connected to the data acquisition and control module, is used to receive energy consumption parameters, convert them into unified standardized energy consumption indicators, and output energy efficiency assessment indicators. The comfort algorithm construction module is connected to the data acquisition and control module to receive the environmental parameters and personnel activity parameters and generate a dynamic comfort index. The association rule mining module is connected to the data acquisition and control module, the comprehensive energy consumption standardization module, and the comfort algorithm construction module. It is used to mine multidimensional association rules between the environmental parameters, standardized energy consumption indicators, and comfort index from historical data. The intelligent decision-making module is connected to the comprehensive energy consumption standardization module, the comfort algorithm construction module, and the association rule mining module, and is used to generate the optimal control strategy based on the standardized energy consumption index, the comfort index, and the association rules. The control command generation module, connected to the intelligent decision-making module and the data acquisition and control module, is used to convert the optimal control strategy into equipment regulation parameters and send them to the data acquisition and control module to drive the cold source system to execute the control commands.

2. The multi-parameter intelligent cold source control system based on big data according to claim 1, characterized in that, The data acquisition and control module includes: Energy metering units are used to monitor the consumption of electricity, natural gas, and water resources; Thermodynamic state monitoring unit is used to monitor the three-dimensional temperature field of the evaporator and condenser, as well as the phase change state of the refrigerant; The environmental sensing unit is used to monitor indoor light intensity, indoor carbon dioxide concentration, and personnel density and distribution.

3. The multi-parameter intelligent cold source control system based on big data according to claim 1, characterized in that, The integrated energy consumption standardization module constructs a unified multi-energy evaluation system to convert heterogeneous energy sources such as electricity, natural gas, and water into standard coal equivalents based on their calorific value. This unified multi-energy evaluation system is based on a dynamic conversion model, which is as follows: , where E represents the total standard coal equivalent, α_i represents the conversion factor for the i-th energy source, and Q_i represents the energy consumption of the i-th energy source.

4. The multi-parameter intelligent cold source control system based on big data according to claim 1, characterized in that, The integrated energy consumption standardization module outputs the following energy efficiency evaluation indicators in real time: , where E_t represents the energy consumption standard coal equivalent at the current moment, and E_{min} represents the historical minimum energy consumption standard coal equivalent.

5. The multi-parameter intelligent cold source control system based on big data according to claim 1, characterized in that, The comfort algorithm construction module calculates the dynamic comfort index based on the PMV-PPD model, and constructs a comfort reward item based on the dynamic comfort index. When the predicted proportion of dissatisfied personnel PPD value is lower than the preset threshold, a positive reward is given; when the predicted proportion of dissatisfied personnel PPD value is higher than the preset threshold, a negative reward is given.

6. The multi-parameter intelligent cold source control system based on big data according to claim 1, characterized in that, The association rule mining module uses an improved FP-Growth algorithm to mine high-frequency association rule sets and extract non-explicit performance-intensive association rules.

7. The multi-parameter intelligent cold source control system based on big data according to claim 1, characterized in that, The intelligent decision-making module uses a deep reinforcement learning algorithm, with energy consumption and PMV comfort index as reward functions, to output compressor frequency and electronic expansion valve opening commands.

8. The multi-parameter intelligent cold source control system based on big data according to claim 7, characterized in that, The reward function R is composed of a weighted sum of the comfort reward term R_comfort and the energy reward term R_energy, as shown in the formula: , where ω_1 and ω_2 are weighting coefficients.

9. A multi-parameter intelligent control method for cold sources based on big data, characterized in that, Including the following steps: Collect energy consumption parameters, environmental parameters, personnel activity parameters, and equipment operating status parameters of the cold source system, and execute control commands; Different types of energy consumption parameters will be uniformly replaced with standardized energy consumption indicators. A dynamic comfort index is generated based on the environmental parameters and personnel activity parameters. Based on historical operating data, we can explore multi-dimensional correlation rules between operating environment parameters, standardized energy consumption indicators, and comfort index. Using the standardized energy consumption index and the comfort index as optimization targets, and combining the association rules, an optimal control strategy is generated through an intelligent learning algorithm. The optimal control strategy is then translated into specific equipment control parameters and executed.

10. The multi-parameter intelligent control method for cold sources based on big data according to claim 9, characterized in that, The intelligent learning algorithm is a deep reinforcement learning algorithm; when the mining association rules trigger a specific emergency situation, an emergency control strategy is activated; the deep reinforcement learning algorithm adopts a priority experience replay mechanism to assign higher weights to the training samples corresponding to the emergency situation.

Images