Intelligent method for energy-saving control and operation and maintenance optimization of refrigerating system based on intelligent algorithm

Abstract

The invention relates to the technical field of intelligent energy-saving control, in particular to a refrigerating system energy-saving control and operation and maintenance optimization intelligent method based on an intelligent algorithm, which comprises the following steps of: acquiring operating parameters of a refrigerating system through remote communication to construct a feature set, evaluating an equipment state based on a fuzzy control algorithm and generating target parameters, deviation between the target and the actual operation state is analyzed, correction parameters are set, the frequency is adjusted, and the cooling liquid temperature is corrected in a linkage mode. According to the method, technical means of cooling tower start-stop control, heat exchange frequency adjustment, intelligent fuzzy control, system integration, equipment detection and the like are fused, and unified datamation expression of various equipment operation characteristics is completed by using operation data collected in real time, so that the key equipment state of the refrigeration system has a traceability and quantifiable management basis; a control instruction is generated through a fuzzy reasoning method in combination with a preset rule under the dynamic working condition, and the recognition and regulation capacity of the equipment operation state under the condition of cooling and heating load fluctuation is enhanced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent energy-saving control technology, and in particular to an intelligent method for energy-saving control and operation and maintenance optimization of a refrigeration system based on an intelligent algorithm. Background Art

[0002] The field of intelligent energy-saving control technology involves research on building energy management and system operation efficiency optimization, mainly including the development of energy-saving control strategies for building energy consumption systems such as cooling, heating, ventilation, and lighting, energy efficiency monitoring technology, load forecasting and regulation methods, equipment operation and maintenance optimization, and coordinated scheduling of multiple energy systems.

[0003] Among them, the energy-saving control and operation and maintenance optimization methods of the refrigeration system refer to the methods commonly used in commercial or industrial buildings to reduce the operating energy consumption of the refrigeration system and improve the system operation and maintenance efficiency, such as set temperature graded control, scheduled start and stop management, parameter adjustment based on empirical rules, manual inspection and planned maintenance.

[0004] Existing technologies generally rely on fixed control strategies to implement energy-saving management for refrigeration systems. These strategies lack the ability to deeply mine operational data from equipment such as cooling towers and chilled water pumps. During operational adjustments, start / stop decisions and frequency settings are primarily guided by manual experience, making it difficult to dynamically respond to load changes and equipment characteristics. Unified logical instructions cannot be formed during the coordinated operation of multiple devices, often leading to operational misalignment or delayed load response. For example, in scenarios with rapidly fluctuating cooling and heating loads, the lack of linkage between cooling tower start / stop sequences and pump frequency adjustment can lead to fluctuations in system cooling capacity or short-term spikes in energy consumption. Furthermore, the lack of a lifecycle-based data analysis mechanism during equipment operation results in inconsistent management strategies for equipment status at different stages. This inability to effectively identify operational hazards such as equipment aging and frequent starts and stops can easily lead to decreased equipment efficiency or unplanned failures. Furthermore, the traditional model lacks a unified system integration platform and the ability to coordinate perception and control between subsystems. This results in delayed energy-saving control and untimely equipment maintenance responses in refrigeration systems, such as the central air conditioning subsystem. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and propose an intelligent method for energy-saving control and operation and maintenance optimization of refrigeration systems based on intelligent algorithms.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: an intelligent method for energy-saving control and operation and maintenance optimization of a refrigeration system based on an intelligent algorithm, comprising the following steps: S1: Obtain the operating parameters of the refrigeration system equipment through the remote communication link and build the equipment operation feature set; S2: Evaluate the current operating status of each type of equipment in the equipment operating feature set using a fuzzy control algorithm, and assign a target operating parameter set to each type of equipment; S3: Analyze the deviation between the target operating parameter set and the current operating state of each type of equipment, and set the operating frequency correction parameter set for each type of equipment according to the deviation type; S4: updating the corresponding operating frequency of each type of equipment according to the operating frequency correction parameter set, calculating the cooling load caused by the change in the updated operating frequency, and adjusting the host coolant temperature setting value in a linked manner to generate a host coolant temperature adjustment result; S5: Based on the host coolant temperature adjustment result, the cloud control platform updates the latest operating parameters of each type of equipment in the equipment operation feature set to obtain the refrigeration system operation and maintenance optimization result.

[0007] As a further solution of the present invention, the equipment operation feature set includes the operating frequency fluctuation amplitude, start-stop cycle characteristics and stage load response value, the target operation parameter set includes the frequency setting target, start-stop control instructions and the identification of the linkage equipment, the operating frequency correction parameter set includes the frequency adjustment amplitude and the adjustment applicable range, the host coolant temperature adjustment result includes the coolant temperature correction value and the set point update status, and the refrigeration system operation and maintenance optimization result includes the frequency update result, the coolant temperature adjustment result and the equipment operation status update identification.

[0008] As a further solution of the present invention, the steps of obtaining the device operation feature set are specifically as follows: S111: Obtain the operating frequency, start / stop status, supply / return water temperature, and operating time of the cooling tower, cooling water pump, chilled water pump, and chiller in the refrigeration system equipment. Extract the corresponding field content based on the communication protocol format of each type of equipment. Combine the equipment code and time information to perform field cleaning and data structure unification to generate a unified format equipment data set. S112: Divide the operating frequency sequence and start / stop state sequence of each device in the unified format device data set into multiple segments according to a time sliding window, extract the operating frequency mean, frequency standard deviation, and start / stop count in each segment, and construct a sequence statistical feature table together with the device identifier and timestamp; S113: Combining the sequence statistical feature tables into a joint expression for device operation features according to time segments, summarizing the combined structure according to device types, and generating a device operation feature set.

[0009] As a further solution of the present invention, the step of obtaining the target operating parameter set is specifically as follows: S211: setting fuzzy control items based on the mean operating frequency, standard deviation of frequency, and number of starts and stops of each type of equipment in the equipment operating characteristic set, constructing a fuzzy membership function, and performing fuzzy reasoning on the input data in combination with a control rule library to generate a fuzzy state membership result; S212: Based on the membership value of each fuzzy control item in the fuzzy state membership result, classify and determine the operating state of each type of equipment according to the maximum membership principle to generate an equipment state control classification result; S213: Generate a target operating parameter set according to the control level and associated control requirements of each device in the device state control classification result.

[0010] As a further solution of the present invention, the step of obtaining the operating frequency correction parameter set is specifically as follows: S311: Obtain the current operating status of each type of equipment, including actual operating frequency and start / stop status information, call the target operating frequency and start / stop control value of the corresponding equipment in the target operating parameter set as a reference benchmark, calculate the deviation rate between the current operating frequency and the target operating frequency, and the time offset between the current start / stop time point and the target control time sequence, and generate an operating status deviation measurement result; S312: Constructing a two-dimensional input variable vector based on the operating frequency deviation rate and the start / stop time offset value of each group of equipment in the operating state deviation measurement result, identifying the deviation type level corresponding to the current operating state of the equipment using a K-nearest neighbor algorithm, and generating a deviation type discrimination result; S313: Call the classification result and deviation level of each type of equipment in the deviation type discrimination result, match the operating frequency correction rule and the start-stop adjustment rule accordingly, set the operating frequency correction value and the start-stop control update instruction, and generate an operating frequency correction parameter set.

[0011] As a further solution of the present invention, the steps of obtaining the host coolant temperature adjustment result are specifically as follows: S411: Calling the operating frequency correction value and device identification information of each type of equipment in the operating frequency correction parameter set, writing the operating frequency correction value into the frequency setting channel of the corresponding frequency-controlled equipment, and synchronously recording the updated frequency and control timing status to generate a device operating frequency update record; S412: Obtaining operating load data corresponding to the water flow rate based on the equipment operating frequency update record, analyzing the instantaneous cooling load change trend caused by the operating frequency change, and generating cooling load response change analysis results; S413: Based on the load change interval in the cooling load response change analysis result and the current host cooling load status, combined with the current coolant temperature set point, derive the coolant temperature adjustment value corresponding to the load change, and superimpose the coolant temperature adjustment value and the current coolant temperature set point to generate a host coolant temperature adjustment result.

[0012] As a further solution of the present invention, the steps for obtaining the refrigeration system operation and maintenance optimization results are specifically as follows: S511: Based on the host coolant temperature adjustment result, the operating frequency setting value, start / stop control command, and coolant temperature adjustment value of each type of equipment are packaged and pushed to the control terminals of the cooling tower, cooling water pump, chilled water pump, and chiller according to the device address mapping relationship, and a remote control command issuance record is generated; S512: Based on the remote control instruction issuance record, collect the latest data on the operating frequency, coolant temperature, and start / stop status of each type of equipment after control, compare the equipment control target with the actual feedback result, and generate an equipment status feedback update data set; S513: Based on the device status feedback, each operating parameter data in the data set is updated, a data structure in a unified format is reconstructed according to the device type and timestamp, and the device operating feature set is synchronously updated to generate a refrigeration system operation and maintenance optimization result.

[0013] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by integrating technical means such as cooling tower start-stop control, heat exchange frequency adjustment, intelligent fuzzy control, system integration and equipment detection, the real-time collected operating data is used to complete the unified digital expression of the operating characteristics of various types of equipment, so that the key equipment status of the refrigeration system has a traceable and quantifiable management basis. Under dynamic working conditions, the fuzzy reasoning method is combined with preset rules to generate control instructions, thereby enhancing the ability to identify and regulate the equipment operating status under fluctuations in cold and hot loads. With the help of automatic frequency adjustment, the start-stop rhythm of the cooling water pump and the cooling tower is flexibly regulated to achieve coordinated debugging and energy consumption optimization among multiple devices. Further, through the closed-loop feedback of data from each subsystem node, the life cycle stage characteristics are accurately extracted and the operation strategy is continuously updated, so that the control logic of the refrigeration system and the equipment status are dynamically adapted. Through the full life cycle management model, energy efficiency and safety assurance capabilities are improved, the system's perception and response capabilities to energy consumption, load changes and equipment aging trends are enhanced, the degree of refinement of debugging management is improved, the frequency of human intervention and debugging costs are reduced, and the simultaneous improvement of equipment operating efficiency and the effective compression of the system debugging cycle are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a schematic diagram of the main steps of the present invention; Figure 2This is a flow chart of step S1 of the present invention; Figure 3 This is a flow chart of step S2 of the present invention; Figure 4 This is a flow chart of step S3 of the present invention; Figure 5 This is a flow chart of step S4 of the present invention; Figure 6 This is a flow chart of step S5 of the present invention. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0016] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0017] See also Figure 1 The present invention provides a technical solution: an intelligent method for energy-saving control and operation and maintenance optimization of a refrigeration system based on an intelligent algorithm, comprising the following steps: S1: Obtain the operating parameters of the refrigeration system equipment through the remote communication link and build the equipment operation feature set; S2: Evaluate the current operating status of each type of equipment in the equipment operating feature set through a fuzzy control algorithm and assign a target operating parameter set to each type of equipment; S3: Analyze the deviation between the target operating parameter set and the current operating status of each type of equipment, and set the operating frequency correction parameter set for each type of equipment according to the deviation type; S4: updating the corresponding operating frequency of each type of equipment according to the operating frequency correction parameter set, calculating the cooling load caused by the change in the updated operating frequency, adjusting the host coolant temperature set value in a linked manner, and generating a host coolant temperature adjustment result; S5: Based on the host coolant temperature adjustment result, the cloud control platform updates the latest operating parameters of each type of equipment in the equipment operation feature set to obtain the cooling system operation and maintenance optimization result; Among them, the equipment operation characteristic set includes the operating frequency fluctuation amplitude, start-stop cycle characteristics and stage load response value, the target operation parameter set includes the frequency setting target, start-stop control instructions and the identification of the linkage equipment, the operating frequency correction parameter set includes the frequency adjustment amplitude and the adjustment applicable range, the host coolant temperature adjustment result includes the coolant temperature correction value and the set point update status, and the refrigeration system operation and maintenance optimization result includes the frequency update result, the coolant temperature adjustment result and the equipment operation status update identification.

[0018] See also Figure 2 , the specific steps for obtaining the device operation feature set are: S111: Obtain the operating frequency, start / stop status, supply / return water temperature, and operating time of the cooling tower, cooling water pump, chilled water pump, and chiller in the refrigeration system equipment. Extract the corresponding field content according to the communication protocol format of each type of equipment. Combine the equipment code and time information to perform field cleaning and data structure unification to generate a unified format equipment data set.

[0019] Obtain the operating frequency, start / stop status, coolant temperature, return water temperature and operating time of the cooling tower, cooling water pump, chiller in the refrigeration system equipment. According to the communication protocol used by each type of equipment, such as BACnet for chillers and ModbusTCP for cooling water pumps, extract the original data according to the field standard in the platform data access module. Assume that a cooling water pump uploads a set of data at 10:00, and the record content includes an operating frequency of 41.7Hz, a start / stop status of 1, a coolant temperature of 31.4Hz, a return water temperature of 35.2℃, and an operating time of 3.75 hours. This data will be combined with the equipment number and upload time to form a standard record entry. For the case where the temperature field unit uploaded by the equipment is one hundredth of a degree Celsius or Fahrenheit, it is necessary to The data is uniformly converted to degrees Celsius and one decimal place is retained. For records with missing fields, the data of the previous time period of the device is called for linear interpolation repair. For example, if the previous period is 40.5Hz and the next period is 41.3Hz, the missing segment is filled with 40.9Hz as the repair value. During the field cleaning process, invalid characters such as "Null", "NaN" or "——" are removed, and a filtering range is set for abnormal values. For example, the operating frequency is limited to 20Hz to 50Hz, and the temperature is limited to 5℃ to 45℃. Data outside this range is directly marked as invalid and not included in the storage. All cleaned fields are uniformly named and structured according to the device code and timestamp to form a unified structure of the equipment operation record data set, ensuring that the data uploaded by different devices is consistent in the field dimensions.

[0020] S112: Based on the operating frequency sequence and start-stop status sequence of each device in the unified format device data set, the data is divided into multiple segments according to the time sliding window. The operating frequency mean, frequency standard deviation and start-stop times in each segment are extracted, and a sequence statistical feature table is constructed together with the device identifier and timestamp.

[0021] According to the operating frequency and start-stop status of each device in the unified format device data set, the time window is set to 30 minutes and the step size is 10 minutes. The continuous segments are divided and the characteristic values are calculated in sequence. In each segment, the average operating frequency is obtained by accumulating the frequency of each time point and dividing it by the frequency. For example, a chilled water pump is sampled six times from 14:00 to 14:30, with frequencies of 39.8Hz, 40.3Hz, 40.1Hz, 40.4Hz, 40.0Hz, and 39.9Hz respectively. The total is 240.5Hz and the average is 40.1Hz. The frequency fluctuation value is calculated by calculating the square of the deviation between each time point and the average value and then taking the root mean square, which is about 0.2. The number of starts and stops is determined by counting the number of times the status field jumps from 0 to 1 or 1 to 0 within the time period. If the device jumps three times in this section, the number of starts and stops is recorded as 3. The above three values are combined with the current time window start time 14:00 and the device number to form a structure field. Each type of equipment generates multiple time period feature records in this way. For example, the average operating frequency of the cooling tower from 15:00 to 15:30 is 34.6Hz, the fluctuation value is 0.1, and the number of starts and stops is 0, indicating that the device is stable during this period, reflecting its stable operating load. All equipment generates statistical results according to the sliding section and merges them to form a time-distributed sequence statistical feature table.

[0022] S113: Combining the sequence statistical feature tables into a joint expression for equipment operation features according to time segments, summarizing the combined structure according to equipment types, and generating an equipment operation feature set.

[0023] The frequency mean, frequency fluctuation value and start-stop times of each type of equipment in each time window are combined into a joint structure of equipment operation characteristics, and organized in chronological order to form a time segment feature expression. For example, the frequency mean of the chiller in the segment from 09:00 to 09:30 is 44.8Hz, the fluctuation value is 0.15, and the start-stop times are 1. In the segment from 09:10 to 09:40, the frequency mean rises to 45.5Hz, the fluctuation value drops to 0.1, and the start-stop times are 0, indicating that the current operating frequency of the equipment is increasing while the fluctuation is decreasing, and it is running continuously. The state remains stable, and the trend changes in the time series are merged to form the expression of the operating stability of the equipment. In addition, the feature fields are grouped by equipment type. The focus of analyzing the operating frequency changes and start-stop frequency of water pump equipment is on the analysis. The cooling tower is mainly classified according to the number of starts and stops. The state feature set of the chiller is constructed based on the frequency stability and the proportion of continuous operation time. The corresponding feature combination structures of all equipment are uniformly organized into a standard feature table, forming an equipment operation feature set with equipment number and time period index as identifiers, and the operating frequency mean, frequency fluctuation value and start-stop number as fields.

[0024] See also Figure 3 , the specific steps for obtaining the target operating parameter set are: S211: Set fuzzy control items according to the operating frequency mean, frequency standard deviation and start-stop times of each type of equipment in the equipment operating characteristic set, construct a fuzzy membership function and perform fuzzy reasoning on the input data in combination with the control rule library to generate a fuzzy state membership result.

[0025] According to the mean operating frequency, standard deviation of frequency and number of starts and stops of each type of equipment in the equipment operation characteristic set, the fuzzy control items are set as "high operating frequency", "violent frequency fluctuation" and "frequent starts and stops", and the Gaussian and S-type membership functions are used to numerically calculate their membership degrees.

[0026] For the first item "operating frequency is too high", record the average operating frequency of the device as , the membership function takes the Gaussian form: ; in, : represents the periodic mean of the equipment operating frequency, in Hz, derived from the statistical results of the equipment characteristic sequence in the previous stage; : Indicates the characteristic center value of "high operating frequency". The value should be based on the set standard value near the upper limit of the equipment's allowable operating frequency. For example, the rated operating frequency of a chiller is usually around 45Hz, and this value is selected as the center of the fuzzy high frequency. : This is the standard deviation control item, reflecting the width of the fuzzy curve. It is set based on the normal fluctuation range of the equipment frequency adjustment range, generally half of the allowable floating range on both sides of the rated frequency; :express The membership degree of the “high operating frequency” language item ranges from 0 to 1.

[0027] If the average operating frequency of a device is ,but: .

[0028] For the second item "violent frequency fluctuation", the standard deviation of the operating frequency is , using the S-type function form: ; in, : Indicates the standard deviation of the operating frequency, used to measure the current frequency stability; : Slope control parameter, which controls the steepness of the transition from low membership to high membership. Its setting is based on the sensitivity response requirements of device data fluctuations. If the system needs to respond more sensitively to fluctuations, the slope should be set larger. : Critical fluctuation value, which is the dividing line for "severe frequency fluctuation". It is calculated based on historical operating data. For example, if it exceeds 0.3Hz, it is considered that the system fluctuation is significant; :express The degree of membership to the language item “frequency fluctuates dramatically”.

[0029] like ,but: .

[0030] For the third item "Frequent start and stop", record the number of start and stop times in the cycle as , using the same form of S-type membership function: ; in, : Indicates the number of starts and stops of a device within the statistical period, which is a discrete integer variable; : Slope control item, used to control the response steepness of the function. The start and stop behavior of the device is greatly affected by frequent switching, so the slope should not be set too high to avoid sudden changes in fuzzy judgment; : is the empirical threshold for frequent starts and stops. Based on industry experience and equipment protection logic, a start and stop frequency exceeding four is generally considered frequent start and stop behavior. : Indicates the degree of membership of the start-stop frequency to the “frequent start-stop” language item.

[0031] like ,but: .

[0032] In summary, the membership results of a certain device in the current cycle are: For "High operating frequency": , for "violent frequency fluctuations": , for “frequent starts and stops”: .

[0033] In the fuzzy control item "operating frequency is too high", the membership value calculated is , indicating that the current average operating frequency of the equipment is in a medium to high state relative to the set high center value, with a certain deviation but not exceeding the standard seriously; in the "Frequency Fluctuation" item, the membership value , indicating that the frequency fluctuation has a clear trend of deviating from stability, but has not yet completely entered the extreme fluctuation range; in the "frequent start and stop" item, the membership value , indicating that the device has been frequently started and stopped during the current cycle, approaching or exceeding the threshold for "frequent starts and stops" in the fuzzy rule. These membership values essentially represent the degree of proximity between each input feature and its fuzzy language term. Values closer to 1 indicate a closer fit to the fuzzy language description.

[0034] During the specific operation process, first, based on the three data of operating frequency mean, frequency standard deviation and start-stop times extracted from the equipment operation feature set, the corresponding fuzzy control items "high operating frequency", "violent frequency fluctuation" and "frequent start-stop" are set respectively. Then, a unique membership function form is constructed for each input, where the operating frequency adopts a Gaussian function to model its symmetrical offset characteristics relative to the center value, and the frequency standard deviation and start-stop times adopt an S-type function to capture their rapid response change characteristics near a specific threshold. Then, the input data is substituted into the corresponding function, and the membership value of each fuzzy control item is calculated in turn. These membership values are then passed into the fuzzy control rule library as input variables, matching the logical attribution conditions of the equipment status in the preset rules, and identifying the fuzzy state set activated by each input through the rules, thereby generating the fuzzy state membership results of each type of equipment in the current operating cycle.

[0035] S212: Based on the membership value of each fuzzy control item in the fuzzy state membership result, the operating state of each type of equipment is classified and determined according to the maximum membership principle to generate an equipment state control classification result.

[0036] Based on the membership values of the three fuzzy control items presented in the fuzzy state membership results, the authors first compared the membership values of 0.546 for high operating frequency, 0.622 for severe frequency fluctuations, and 0.817 for frequent starts and stops. The maximum value of 0.817 was selected as the dominant attribute of the equipment's operating state in the current cycle, indicating that the equipment's operating behavior is most closely aligned with the "frequent starts and stops" control item, thus triggering the priority activation of the start-stop-related rules. This "maximum membership determination" method, derived from the membership competition strategy in fuzzy reasoning, explicitly stipulates that when multiple skewed features are present, the item with the strongest response is used as the benchmark for identifying the equipment's current operating state. Subsequently, the system calls the classification mapping relationship in the control rule library, corresponds the frequent start and stop items to the "start and stop regulation dominant type" state, and completes the classification mapping in combination with the equipment type. For example, the cooling water pump is classified as the "high-frequency load response type" in this state, and the cooling tower is classified as the "cycle overload waiting type". It is assumed that the current equipment is marked as the first-level response level because the membership of this item is higher than the preset threshold value. The setting basis of the preset threshold value is mainly based on the combination of historical operation data statistical analysis and expert experience rules. By comparing the membership distribution characteristics of a large number of equipment in normal and abnormal states, the various fuzzy control items are extracted in terms of their performance. The typical interval of membership when there is obvious skewed behavior is used to determine the critical threshold range; at the same time, combined with the equipment response sensitivity and false alarm tolerance requirements in actual projects, reasonable intervention trigger standards are set to ensure that when the membership of a fuzzy item exceeds the threshold, the abnormal feature it represents has sufficient discrimination strength, thereby having a logical basis for triggering the priority control rule, avoiding the system's excessive response to slight fluctuations, and achieving a balance between sensitivity and robustness. The system identifies it as a significant abnormal feature and lists it as a priority intervention object, and records the equipment number, dominant control item identifier, classification label and control level number into a structure, which is output as the equipment status control classification result.

[0037] S213: Generate a target operating parameter set according to the control level and associated control requirements of each device in the device state control classification result.

[0038] Based on the "start / stop regulation-dominated" status label and the first-level control level marker in the device state control classification results, the system automatically calls the corresponding instruction template from the instruction template library. This instruction template is set by the control rule designer based on the device response model and empirical rules. It specifies that at the first level, the start / stop logic needs to be directly modified. The operating frequency is also associated as a collaborative variable to enhance the regulation response capability. For example, at this level, a chilled water pump must implement a two-dimensional regulation of a 3-minute start / stop delay and an 8-minute frequency increase. For chillers, a host cooling capacity calibration flag is added for subsequent temperature control linkage. The setting range in the template is derived from the device's historical operating stability boundaries, fault warning behaviors, and energy-saving strategy association logic. By replacing each policy item in the template and associating it with the current actual device state parameters, the three target parameters of the operating frequency target value, the start / stop control instruction value, and the linkage control flag are determined. These values are then aggregated and encapsulated into a standardized control structure, which is bound to the device code and current cycle identifier as unique record units to generate a target operating parameter set with a complete description of the control behavior.

[0039] See also Figure 4 , the specific steps for obtaining the operating frequency correction parameter set are: S311: Obtain the current operating status of each type of equipment, including the actual operating frequency and start-stop status information, call the target operating frequency and start-stop control value of the corresponding equipment in the target operating parameter set as a reference benchmark, calculate the deviation rate between the current operating frequency and the target operating frequency, and the time offset between the current start-stop time point and the target control timing, and generate the operating status deviation measurement result.

[0040] To obtain the current operating status of each device type, including the actual operating frequency and start / stop status information, the frequency sampling values and start / stop status flags in the device's real-time operating records are retrieved and field-concatenated with the target operating parameter set generated in the previous control cycle. Each device group is mapped one-to-one by device number. The target operating frequency and start / stop control value are extracted as a benchmark. The deviation rate between the actual operating frequency and the target frequency is then calculated. This calculation process uses the target value as the reference point, and the difference is normalized with the target value, retaining two decimal places. In this example, the current operating frequency of a cooling water pump is 41. 2Hz, the target frequency is 45.0Hz, then the deviation rate is negative 8.4%, indicating that the current frequency is lower than the target value. For the calculation of the start and stop time points, the last start and stop switching timestamp of the current device is called, and the time difference is compared with the target start and stop timing setting value to obtain the time offset result. In this example, if the device is scheduled to start and stop at 13:30 and actually starts and stops at 13:42, the offset time is 12 minutes, indicating a delay in the control response. After all devices complete the calculation of the frequency deviation rate and time offset value, each group of calculation results is structured and stored to generate the operating status deviation calculation result.

[0041] S312: Based on the operating frequency deviation rate and start-stop time offset value of each group of equipment in the operating status deviation measurement results, a two-dimensional input variable vector is constructed, and the deviation type level corresponding to the current operating status of the equipment is identified through the K-nearest neighbor algorithm to generate a deviation type judgment result.

[0042] First, set the two component variables in the input vector as follows: : Indicates the operating frequency deviation rate, expressed in percentage. It is calculated by dividing the difference between the actual operating frequency of the device and the target frequency by the target frequency, and then multiplying by 100. It indicates the degree of frequency deviation. : Indicates the start / stop time offset value, in minutes, which represents the time difference between the actual start / stop time of the device and the planned start / stop time. A positive value indicates a delayed response, while a negative value indicates an early response. : Constitutes the operating status input vector of the device to be determined in this cycle.

[0043] Take a set of example data to illustrate that the current frequency of a device is 42.0Hz and the target frequency is 45.0Hz. Then: ; The planned start and stop time is 10:00, but it actually occurs at 10:08. Then: ; Therefore, the input vector of the device in the current cycle is .

[0044] Then, several labeled historical deviation type samples are extracted from the preset training sample set as reference points. For example, samples A, B, and C correspond to different operating status types respectively. The Euclidean distance between the current vector and each training sample is calculated. The distance calculation formula is: ; in: 、 : It is the deviation input variable of the device to be determined; :For the training sample Two deviation attributes of the data; : Represents the distance measure between the current input and the th sample.

[0045] Assume sample A is , B is , C is , corresponding to the three deviation types of "insufficient frequency type", "start-stop hysteresis type" and "bidirectional deviation type", and calculated in sequence: ; ; .

[0046] From the above results, it can be seen that sample A is closest to the current device and is therefore identified as the current state type.

[0047] The value of K determines the classification granularity, and an odd number is usually used to avoid vote ties. Here, K=1 is used, which means that the nearest neighbor method is used, and only the closest sample is considered. This is suitable for industrial system control scenarios with clear samples and clear type label boundaries. If expanded to K=3, the three closest samples need to be compared and voted on. K=1 is used in the current task because the operating frequency deviation rate and time offset value are both continuously measurable and the sample distribution boundaries are clear. The device is classified as "insufficient frequency type", which means that the current operating frequency of the device is significantly lower than the target value. Although there is a certain lag in starting and stopping, the main problem is the frequency control deviation.

[0048] S313: Call the classification result and deviation level of each type of equipment in the deviation type discrimination result, match the operating frequency correction rule and the start-stop adjustment rule accordingly, set the operating frequency correction value and the start-stop control update instruction, and generate the operating frequency correction parameter set.

[0049] The classification label and corresponding deviation level of each type of equipment in the deviation type identification result are called, and the preset operating frequency correction rule and start-stop adjustment rule template are first matched. The preset operating frequency correction rule and start-stop adjustment rule template are mainly set by integrating historical operating data analysis, expert experience library and multi-scenario simulation test results. The system first extracts typical adjustment response modes under different deviation types based on large sample equipment operation deviation and correction feedback data to form a preliminary rule framework, and then introduces the experience parameters of professional operation and maintenance personnel for manual calibration to ensure the engineering applicability and operability of the rules; on this basis, the control simulation platform is used to repeatedly test and optimize various correction strategies under different operating loads, environmental conditions and equipment combinations. For equipment judged to be "insufficient frequency type", the system reads the corresponding frequency correction strategy parameters. This type generally requires an increase in the current frequency setting value, and the adjustment amplitude is set according to the deviation level. For example, the first-level deviation level corresponds to an increase of 10, and the second-level deviation corresponds to an increase of 5. If the equipment is judged to be "start-stop hysteresis type", the adjustment Using a start / stop logic correction template, the planned start / stop points are rolled back based on the delay time, with the rollback increment increasing by minutes. For devices with "bidirectional deviation," both the frequency and start / stop rules are extracted for coordinated correction, generating frequency adjustment values and start / stop time correction instructions. For different device types, pumps primarily use operating frequency correction, with the adjustment values directly written to the inverter interface. Cooling towers primarily use start / stop correction, resulting in an earlier or later start time. Some cooling towers with fan variable frequency control also add fan frequency correction values. For chillers, if the deviation is identified as "overfrequency," the frequency is reduced. A minimum interval is set in the start / stop control strategy to prevent frequent starts and stops from causing overload. If a host is identified as "early start / stop," the correction instruction postpones the control trigger point and adds a coolant temperature non-linkage tag for subsequent temperature policy filtering. The system encapsulates each device's device number, deviation type, correction value, adjustment instruction, and application tag into a structured control parameter unit, forming an operating frequency correction parameter set.

[0050] See also Figure 5 , the specific steps for obtaining the host coolant temperature adjustment result are: S411: Call the operating frequency correction value and device identification information of each type of equipment in the operating frequency correction parameter set, write the operating frequency correction value into the frequency setting channel of the corresponding frequency-controlled equipment, and synchronously record the updated frequency and control timing status to generate a device operating frequency update record.

[0051] The operating frequency correction value and device identification information of each type of equipment in the operating frequency correction parameter set are called, and the type of control object and the frequency conversion control interface channel are identified according to the equipment number. For example, cooling water pumps, chilled water pumps and cooling tower fans with frequency conversion function all receive frequency setting value updates through the setting interface. The system writes the operating frequency correction value in the correction parameter set to the corresponding channel, overwriting the current frequency setting point. During the control instruction issuance process, the writing time and execution feedback status are recorded at the same time for subsequent tracking of the real-time response of the frequency adjustment instruction. After the parameter writing is completed, the system re-collects the current operating frequency status of the equipment and compares it with the frequency value before the update to confirm that the frequency correction action has taken effect, and records the time point of the action as the anchor point for the control timing change. All frequency update values are encapsulated together with the corresponding timestamp, equipment number, and instruction execution feedback to form an equipment operating frequency update record.

[0052] S412: According to the equipment operation frequency update record, the operation load data corresponding to the water flow is obtained, the instantaneous cooling load change trend caused by the operation frequency change is analyzed, and the cooling load response change analysis result is generated.

[0053] Based on the equipment operating frequency update record, the system obtains the water flow data corresponding to two consecutive time points before and after the frequency adjustment, and calls the conversion relationship based on the equipment type to convert the water flow into the current system operating load data. For example, there is a direct water volume-load coupling relationship between the chilled water pump and the chiller. By monitoring the changing trend of the water flow after the frequency increases, the instantaneous cooling load adjustment direction caused by the frequency change is analyzed. If the frequency increase causes the water flow per unit time to increase, it can be deduced that the load curve is shifted upward accordingly. This shift interval is compared with the load response model during the operating period to determine whether the current frequency change brings about an effective increase in the cooling load. If the system load maintains a stable growth trend in the two cycles after the frequency adjustment, it is marked as a positive response. If there is a load lag or decrease, it is marked as a reverse response or over-modulation behavior. Combined with the frequency change amplitude and the load change direction mark, the cooling load response change analysis results are generated.

[0054] S413: Based on the load change interval in the cooling load response change analysis result and the current host cooling load state, combined with the current coolant temperature set point, derive the coolant temperature adjustment value corresponding to the load change, and superimpose the coolant temperature adjustment value and the current set point to generate the host coolant temperature adjustment result.

[0055] The adjustment calculation is based on a linear fit relationship between load change and coolant temperature, expressed as follows: ; in, : Coolant temperature adjustment value, in degrees Celsius (℃), indicating the extent to which the coolant temperature should change in the current cycle; : Cooling load change value, in kilowatts (kW), refers to the actual increase or decrease in the host cooling load detected by the system before and after frequency adjustment; : Load temperature response coefficient, in °C / kW, used to express the adjustment range of the coolant temperature set value due to unit load change.

[0056] about The setting basis is: This coefficient is derived from the statistical regression results of the host cooling performance curve and operating data. In engineering practice, the host coolant temperature adjustment is usually approximately linearly related to its cooling load, and the response slope is The size of the temperature is determined by the following factors: 1. Host type and energy efficiency level: The load adjustment capabilities of screw and water-cooled centrifugal hosts are different. The higher the energy efficiency, the greater the sensitivity of temperature to load. 2. Historical operation data fitting: By selecting sample points of load changes and corresponding coolant temperature changes in multiple operation cycles, a set of average slopes are fitted using the linear regression method as the basic response coefficient; 3. System design temperature difference range: If the main unit is designed to have a supply and return water temperature difference of 5°C and the corresponding adjustment load range is 150kW to 250kW, the unit adjustment step size can be calculated as ; 4. Adjustment sensitivity requirements: If the system needs to adjust slowly, the response coefficient will be further reduced; if a fast response to load fluctuations is required, a larger negative coefficient will be selected to reflect the slope of the strong adjustment curve.

[0057] During the setup process, multiple devices are usually The value is set to the experience range , and establish boundary conditions in combination with operational stability to avoid system oscillation caused by excessive single temperature adjustment.

[0058] Assume the current coolant temperature set point is , the host load increases from 180kW to 195kW, and we get: ; The selected response coefficient is , then the temperature adjustment value is: , coolant temperature adjustment results for: .

[0059] This indicates that the current coolant temperature should be reduced from 7.0°C to 6.55°C to accommodate the increased cooling load. This adjustment value is written to the cooling system and bound to the coolant temperature setpoint of the host refrigeration circuit. In a multi-host control system, a unified temperature setting can be synthesized based on the primary load unit or average weight, serving as the temperature target for the entire cluster control strategy.

[0060] See also Figure 6 ,The specific steps for obtaining the refrigeration system operation and maintenance optimization results are as follows: S511: Based on the host coolant temperature adjustment result, the operating frequency setting value, start-stop control instructions and coolant temperature adjustment value of each type of equipment are encapsulated and pushed to the control terminals of the cooling tower, cooling water pump, chilled water pump and chiller according to the device address mapping relationship, and a remote control instruction issuance record is generated.

[0061] Based on the host coolant temperature adjustment result, the system encapsulates the operating parameters required to be executed by each type of equipment in the current cycle, including the operating frequency setting value, start and stop control instructions and coolant temperature adjustment value. When generating the encapsulated data, the system calls the operating frequency correction parameter set and the temperature adjustment result to perform field combination, and adds the device type, device address and control cycle number. Then, according to the device address mapping relationship registered in the control system, the encapsulated instructions are pushed to the control terminals of the cooling tower, cooling water pump, chilled water pump and chiller one by one. During the issuance process, the system records the transmission duration and response status of each instruction data packet, and marks whether it is successfully written to the target control interface. For example, the target frequency of the cooling water pump is set to 42.5Hz, the coolant temperature setting of the chiller is adjusted to 6.6℃, and the start and stop status of the cooling tower remains unchanged. All instructions and their execution addresses, response times and control contents are uniformly constructed into an instruction issuance record table to form a remote control instruction issuance record.

[0062] S512: Based on the records issued by the remote control instructions, the latest data on the operating frequency, coolant temperature, and start / stop status of each type of equipment after control is collected, the equipment control target is compared with the actual feedback results, and an equipment status feedback update data set is generated.

[0063] Based on the control execution feedback results in the remote control command issuance record, the system calls the operating status interface of each control terminal to collect the latest data on frequency, coolant temperature, and start / stop status, and performs field-level comparison with the previous control target to extract the response accuracy difference of the execution result. In the frequency control part, if the set value of the chilled water pump is 44.0Hz and the actual collected value is 43.7Hz, the recording error is 0.3Hz. In the coolant temperature part, if the target adjustment value of the chiller is 6.5℃ and the actual value is 6.4℃, the temperature difference is 0.1℃. In the start / stop status part, if the control target is on and the actual status value feedback is 1, it is determined that the command is fully executed successfully. If the return value is 0, it is marked as an unresponsive device. All differences and matching flags are uniformly written into the feedback status structure table. The data contains the device number, parameter category, set value, feedback value, and response identifier to form a device status feedback update data set.

[0064] S513: Based on the equipment status feedback, each operating parameter data in the data set is updated, and a unified data structure is reconstructed according to the equipment type and timestamp. The equipment operation feature set is also updated synchronously to generate the refrigeration system operation and maintenance optimization results.

[0065] Based on the equipment status feedback, each operating parameter data in the data set is updated. First, the corresponding field structure is extracted according to the equipment type, and the data structure in a unified format is reconstructed in combination with the timestamp field. For variable frequency or start-stop equipment such as cooling water pumps, chilled water pumps and cooling towers, the system reorganizes the operating frequency, start-stop status and response period as the main fields and generates standard record entries. For chillers, the coolant temperature and load correction paragraphs are written synchronously, and at the same time, it is determined whether each field has changed. If the coolant temperature of the chiller is set to 6.7℃ in the previous cycle and 6.5℃ in the current cycle, the temperature difference is recorded as 0.2℃, which is marked as a valid correction. If the start-stop status of the cooling tower remains at 1 within two cycles, it does not constitute an operating feature update. The system writes the valid field change value to the update identifier and synchronously refreshes the original data records of the corresponding equipment and time period in the equipment operation feature set to form the refrigeration system operation and maintenance optimization results.

[0066] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. An intelligent method for energy-saving control and operation and maintenance optimization of refrigeration systems based on intelligent algorithms, characterized by: The following steps are involved: S1: Obtain the operating parameters of the refrigeration system equipment through the remote communication link and build the equipment operation feature set; S2: Evaluate the current operating status of each type of equipment in the equipment operating feature set using a fuzzy control algorithm, and assign a target operating parameter set to each type of equipment; S3: Analyze the deviation between the target operating parameter set and the current operating state of each type of equipment, and set the operating frequency correction parameter set for each type of equipment according to the deviation type; S4: updating the corresponding operating frequency of each type of equipment according to the operating frequency correction parameter set, calculating the cooling load caused by the change in the updated operating frequency, and adjusting the host coolant temperature setting value in a linked manner to generate a host coolant temperature adjustment result; S5: Based on the host coolant temperature adjustment result, the cloud control platform updates the latest operating parameters of each type of equipment in the equipment operation feature set to obtain the refrigeration system operation and maintenance optimization result.

2. The intelligent method for energy-saving control and operation and maintenance optimization of refrigeration systems based on intelligent algorithms according to claim 1 is characterized in that: The equipment operation feature set includes the operating frequency fluctuation amplitude, start-stop cycle characteristics and stage load response value; the target operation parameter set includes the frequency setting target, start-stop control instructions and the identification of the linkage equipment; the operating frequency correction parameter set includes the frequency adjustment amplitude and the adjustment applicable range; the host coolant temperature adjustment result includes the coolant temperature correction value and the set point update status; the refrigeration system operation and maintenance optimization result includes the frequency update result, the coolant temperature adjustment result and the equipment operation status update identification.

3. The intelligent method for energy-saving control and operation and maintenance optimization of refrigeration systems based on intelligent algorithms according to claim 1 is characterized in that: The steps for obtaining the device operation feature set are specifically as follows: S111: Obtain the operating frequency, start / stop status, supply / return water temperature, and operating time of the cooling tower, cooling water pump, chilled water pump, and chiller in the refrigeration system equipment. Extract the corresponding field content based on the communication protocol format of each type of equipment. Combine the equipment code and time information to perform field cleaning and data structure unification to generate a unified format equipment data set. S112: Divide the operating frequency sequence and start / stop state sequence of each device in the unified format device data set into multiple segments according to a time sliding window, extract the operating frequency mean, frequency standard deviation, and start / stop count in each segment, and construct a sequence statistical feature table together with the device identifier and timestamp; S113: Combining the sequence statistical feature tables into a joint expression for device operation features according to time segments, summarizing the combined structure according to device types, and generating a device operation feature set.

4. The intelligent method for energy-saving control and operation and maintenance optimization of a refrigeration system based on an intelligent algorithm according to claim 3 is characterized in that: The steps for obtaining the target operating parameter set are specifically as follows: S211: setting fuzzy control items based on the mean operating frequency, standard deviation of frequency, and number of starts and stops of each type of equipment in the equipment operating characteristic set, constructing a fuzzy membership function, and performing fuzzy reasoning on the input data in combination with a control rule library to generate a fuzzy state membership result; S212: Based on the membership value of each fuzzy control item in the fuzzy state membership result, classify and determine the operating state of each type of equipment according to the maximum membership principle to generate an equipment state control classification result; S213: Generate a target operating parameter set according to the control level and associated control requirements of each device in the device state control classification result.

5. The intelligent method for energy-saving control and operation and maintenance optimization of a refrigeration system based on an intelligent algorithm according to claim 4 is characterized in that: The steps for obtaining the operating frequency correction parameter set are specifically as follows: S311: Obtain the current operating status of each type of equipment, including actual operating frequency and start / stop status information, call the target operating frequency and start / stop control value of the corresponding equipment in the target operating parameter set as a reference benchmark, calculate the deviation rate between the current operating frequency and the target operating frequency, and the time offset between the current start / stop time point and the target control time sequence, and generate an operating status deviation measurement result; S312: Constructing a two-dimensional input variable vector based on the operating frequency deviation rate and the start / stop time offset value of each group of equipment in the operating state deviation measurement result, identifying the deviation type level corresponding to the current operating state of the equipment using a K-nearest neighbor algorithm, and generating a deviation type discrimination result; S313: Call the classification result and deviation level of each type of equipment in the deviation type discrimination result, match the operating frequency correction rule and the start-stop adjustment rule accordingly, set the operating frequency correction value and the start-stop control update instruction, and generate an operating frequency correction parameter set.

6. The intelligent method for energy-saving control and operation and maintenance optimization of a refrigeration system based on an intelligent algorithm according to claim 5 is characterized in that: The steps for obtaining the host coolant temperature adjustment result are specifically as follows: S411: Calling the operating frequency correction value and device identification information of each type of equipment in the operating frequency correction parameter set, writing the operating frequency correction value into the frequency setting channel of the corresponding frequency-controlled equipment, and synchronously recording the updated frequency and control timing status to generate a device operating frequency update record; S412: Obtaining operating load data corresponding to the water flow rate based on the equipment operating frequency update record, analyzing the instantaneous cooling load change trend caused by the operating frequency change, and generating cooling load response change analysis results; S413: Based on the load change interval in the cooling load response change analysis result and the current host cooling load status, combined with the current coolant temperature set point, derive the coolant temperature adjustment value corresponding to the load change, and superimpose the coolant temperature adjustment value and the current coolant temperature set point to generate a host coolant temperature adjustment result.

7. The intelligent method for energy-saving control and operation and maintenance optimization of a refrigeration system based on an intelligent algorithm according to claim 6 is characterized in that: The steps for obtaining the refrigeration system operation and maintenance optimization results are specifically as follows: S511: Based on the host coolant temperature adjustment result, the operating frequency setting value, start / stop control command, and coolant temperature adjustment value of each type of equipment are packaged and pushed to the control terminals of the cooling tower, cooling water pump, chilled water pump, and chiller according to the device address mapping relationship, and a remote control command issuance record is generated; S512: Based on the remote control instruction issuance record, collect the latest data on the operating frequency, coolant temperature, and start / stop status of each type of equipment after control, compare the equipment control target with the actual feedback result, and generate an equipment status feedback update data set; S513: Based on the device status feedback, each operating parameter data in the data set is updated, a data structure in a unified format is reconstructed according to the device type and timestamp, and the device operating feature set is synchronously updated to generate a refrigeration system operation and maintenance optimization result.

Images