A method and system for diagnosing energy consumption of a compressed air system in a thermal power plant

Abstract

The application relates to the technical field of compressed air systems, in particular to a method and system for diagnosing energy consumption of a compressed air system of a thermal power plant. The method comprises the following steps: establishing a data sensing array of the compressed air system; obtaining operation monitoring data based on the data sensing array, and determining whether to generate an early warning instruction according to the operation monitoring data and a preset abnormal early warning model; generating an energy consumption diagnosis result according to the early warning instruction and a preset energy consumption diagnosis model; through the data sensing array, the operation data of the compressed air system is collected, the energy consumption deviation state of the compressed air system is monitored in real time, and through the double diagnosis mechanism of the added abnormal early warning model and energy consumption diagnosis model, the diagnosis accuracy of the compressed air system of the thermal power plant is improved. The abnormal early warning model performs initial screening on the energy consumption problem according to the real-time energy consumption deviation state, dynamically adjusts the diagnosis strategy according to the screening result in combination with the energy consumption diagnosis model, quickly diagnoses the energy consumption abnormal problem, and improves the energy consumption diagnosis efficiency of the compressed air system of the thermal power plant.

Description

Technical Field

[0001] This application relates to the field of compressed air system technology, and in particular to a method and system for diagnosing the energy consumption of compressed air systems in thermal power plants. Background Technology

[0002] In actual operation of thermal power plants, compressed air systems generally exhibit a management tendency of prioritizing safety over economy. Existing monitoring methods are typically limited to monitoring the main pipeline pressure and the start-up and shutdown status of individual air compressors. Data collection points are scattered and isolated, making it difficult to form a comprehensive understanding of pipeline pressure distribution, equipment matching efficiency, and leakage losses. When system energy consumption abnormally increases, maintenance personnel often face the following technical challenges: Anomaly detection is delayed. Traditional monitoring methods rely on fixed threshold alarms (such as lower pressure limits), which can only trigger alarms when a fault occurs or parameters are severely exceeded. They cannot detect gradual energy efficiency degradation caused by equipment performance decline, micro-leaks in the pipeline network, etc. This reactive response mode results in a large amount of compressed air being wasted unnoticed.

[0003] Diagnosis and localization are difficult. Because compressed air systems involve air compressors, dryers, filters, air tanks, and complex piping networks, different faults (such as adsorption dryer failure, pipeline leaks, and blockages in air-consuming equipment) may exhibit similar characteristics in energy efficiency indicators. Current technologies lack the ability to model specific abnormal scenarios. When energy consumption data fluctuates, maintenance personnel need to investigate the entire system one by one, which is time-consuming and highly dependent on personal experience. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for diagnosing energy consumption in compressed air systems of thermal power plants in order to solve the above-mentioned technical problems, thereby improving the efficiency and accuracy of energy consumption diagnosis in compressed air systems of thermal power plants.

[0005] In some embodiments of this application, by adding a data sensing array, the operating data of the compressed air system is collected, the energy consumption deviation of the compressed air system is monitored in real time, and the diagnostic accuracy of the compressed air system of the thermal power plant is improved by adding a dual diagnostic mechanism of an anomaly early warning model and an energy consumption diagnostic model.

[0006] In some embodiments of this application, the abnormal early warning model performs initial screening of energy consumption problems based on real-time energy consumption deviation status, and dynamically adjusts the diagnostic strategy based on the screening results and the energy consumption diagnosis model to quickly diagnose abnormal energy consumption problems and improve the energy consumption diagnosis efficiency of the compressed air system of thermal power plants.

[0007] In some embodiments of this application, a method for diagnosing the energy consumption of a compressed air system in a thermal power plant is provided, including: Establish a data sensing array for the compressed air system; Acquire operation monitoring data based on the data sensing array, and determine whether to generate a warning instruction according to the operation monitoring data and a preset abnormal warning model; Generate an energy consumption diagnosis result according to the warning instruction and a preset energy consumption diagnosis model; The data sensing array includes a plurality of monitoring points.

[0008] In some embodiments of the present application, the preset abnormal warning model comprises: Acquire historical operation data of the compressed air system, and set a plurality of energy efficiency indicators according to the historical operation data; Establish an initial abnormal warning model according to all the energy efficiency indicators, and set a plurality of abnormal scenarios; Establish an abnormal scenario sequence A, where A=(a1, a2...ai...an), ai is the i-th abnormal scenario, and n is the number of abnormal scenarios; Set ai as the target scenario in sequence according to the abnormal scenario sequence A; Establish a recognition sub-model for the target scenario; Set the recognition sub-model of each abnormal scenario in sequence, and construct an abnormality recognition model according to all the recognition sub-models; Construct the abnormal warning model according to the initial abnormal warning model and the abnormality recognition model.

[0009] In some embodiments of the present application, the establishing a recognition sub-model for the target scenario comprises: Establish a basic indicator set, where the basic indicator set includes a plurality of basic indicators; Generate an associated data packet of the target scenario according to the historical operation data; Set the energy efficiency mapping state of the target scenario according to the associated data packet, where the energy efficiency mapping state includes the expected deviation value of each energy efficiency indicator; Generate the mapping value of each basic indicator to the target scenario; Preset a mapping value threshold H; If H<hi (i=1, 2...r), set the i-th basic indicator as the mapping indicator of the target scenario; where hi is the mapping value of the i-th basic indicator to the target scenario, and r is the number of basic indicators; Acquire all mapping indicators of the target scenario; Set the basic mapping state of the target scenario according to all the mapping indicators; Establish the recognition sub-model of the target scenario according to the energy efficiency mapping state and the basic mapping state.

[0010] In some embodiments of the present application, the determining whether to generate a warning instruction comprises: Preset a plurality of feedback time nodes; Acquire operation monitoring data obtained by the data sensing array at the current monitoring time node; Generate an energy efficiency abnormal value f according to the operation monitoring data; f= si]; where θ1 is the number of energy efficiency indicators; si is the deviation value of the i-th energy efficiency indicator generated according to the operation monitoring data; Preset an energy efficiency abnormal value threshold F1; If f>F1, generate an early warning instruction at the current feedback time node.

[0011] In some embodiments of the present application, the early warning instruction includes: Establish an initial operation state according to the deviation value of each energy efficiency indicator; Generate a first-level similarity value between the initial operation state and each abnormal scenario; Preset a first-level similarity value threshold K1; If K1<ki, (i=1, 2...n), set the i-th abnormal scenario as a to-be-judged scenario; where ki is the first-level similarity value between the initial operation state and the i-th abnormal scenario; n is the number of abnormal scenarios; Set an auxiliary collection instruction according to all the to-be-judged scenarios, and acquire an auxiliary data packet according to the auxiliary collection instruction.

[0012] In some embodiments of the present application, the preset energy consumption diagnosis model includes: Set ai as the to-be-diagnosed scenario in sequence according to the abnormal scenario sequence A; Set a diagnosis sub-strategy for the to-be-diagnosed scenario, where the diagnosis sub-strategy includes a plurality of judgment sub-components; Set the diagnosis sub-strategy for each abnormal scenario in sequence; Construct the energy consumption diagnosis model according to all the diagnosis sub-strategies.

[0013] In some embodiments of the present application, the generating an energy consumption diagnosis result includes: Generate a second-level similarity value with each to-be-judged scenario according to all the auxiliary data packets; Set a first-level diagnosis order according to all the second-level similarity values; Select an execution scenario in sequence based on the first-level diagnosis order; Acquire diagnosis feedback data according to the diagnosis sub-strategy of the execution scenario, and generate a fitness value w according to the diagnosis feedback data; Preset a fitness value threshold W1; If w>W1, generate a first-level diagnosis instruction for the execution scenario; Acquire all the first-level diagnosis instructions, and generate a diagnosis result.

[0014] In some embodiments of the present application, an energy consumption diagnosis system for a compressed air system of a thermal power plant is provided, comprising: a monitoring unit, configured to establish a data sensing array of the compressed air system; where the data sensing array includes a plurality of monitoring points; the monitoring unit includes a plurality of collection sub-modules, and the collection sub-modules are disposed at each monitoring point; a central control unit, configured to acquire operation monitoring data through the data sensing array; the central control unit includes: a first processing module, configured to establish an abnormal early warning model; the first processing module is further configured to determine whether to generate an early warning instruction according to the operation monitoring data and the abnormal early warning model; a second processing module, configured to establish an energy consumption diagnosis model; the second processing module is further configured to generate an energy consumption diagnosis result according to the early warning instruction and the energy consumption diagnosis model.

[0015] In some embodiments of the present application, the first processing module is further configured to: acquire historical operation data of the compressed air system, and set a plurality of energy efficiency indicators according to the historical operation data; establish an initial abnormal early warning model according to all the energy efficiency indicators, and set a plurality of abnormal scenarios; establish an abnormal scenario sequence A, where A=(a1, a2…ai…an), ai is the i-th abnormal scenario; and n is the number of abnormal scenarios; sequentially set ai as a target scenario according to the abnormal scenario sequence A; establish a recognition sub-model for the target scenario; sequentially set the recognition sub-model for each abnormal scenario, and construct an abnormal recognition model according to all the recognition sub-models; construct the abnormal early warning model according to the initial abnormal early warning model and the abnormal recognition model; wherein establishing the recognition sub-model for the target scenario comprises: establish a basic indicator set, where the basic indicator set includes a plurality of basic indicators; generate an associated data packet of the target scenario according to the historical operation data; set the energy efficiency mapping state of the target scenario according to the associated data packet, where the energy efficiency mapping state includes the expected deviation value of each energy efficiency indicator; generate a mapping value of each basic indicator for the target scenario; preset a mapping value threshold H; if H<hi (i=1, 2…r), set the i-th basic indicator as the mapping indicator of the target scenario; where hi is the mapping value of the i-th basic indicator to the target scenario; r is the number of basic indicators; obtain all mapping indicators of the target scenario; set the basic mapping state of the target scenario according to all the mapping indicators; establish a recognition sub-model for the target scenario according to the energy efficiency mapping state and the basic mapping state.

[0016] In some embodiments of the present application, the first processing module is further configured to: preset a plurality of feedback time nodes; obtain the operation monitoring data acquired by the data sensing array at the current monitoring time node; generate an energy efficiency outlier f according to the operation monitoring data; f= si]; where θ1 is the number of energy efficiency indicators; si is the deviation value of the i-th energy efficiency indicator generated according to the operation monitoring data; preset an energy efficiency outlier threshold F1; if f>F1, generate an early warning instruction at the current feedback time node; where the early warning instruction includes: establish an initial operation state according to the deviation value of each energy efficiency indicator; generate a first-level similarity value between the initial operation state and each abnormal scenario; preset a first-level similarity value threshold K1; if K1<ki, (i=1, 2...n), set the i-th abnormal scenario as a scenario to be judged; where ki is the first-level similarity value between the initial operation state and the i-th abnormal scenario; n is the number of abnormal scenarios; set an auxiliary collection instruction according to all the scenarios to be judged, and acquire an auxiliary data packet according to the auxiliary collection instruction.

[0017] Compared with the prior art, the energy consumption diagnosis method and system for a compressed air system of a thermal power plant according to the embodiments of the present application have the following beneficial effects: By adding a data sensing array to collect the operation data of the compressed air system, the energy consumption deviation state of the compressed air system is monitored in real time, and the dual diagnosis mechanism of the abnormal early warning model and the energy consumption diagnosis model is added, so as to improve the diagnosis accuracy of the compressed air system of the thermal power plant.

[0018] The abnormal early warning model performs initial screening on energy consumption problems according to the real-time energy consumption deviation state, dynamically adjusts the diagnosis strategy in combination with the energy consumption diagnosis model according to the screening result, quickly diagnoses energy consumption abnormal problems, and improves the energy consumption diagnosis efficiency for the compressed air system of the thermal power plant. Brief Description of the Drawings

[0019] Figure 1 This is a flowchart illustrating a preferred embodiment of an energy consumption diagnosis method for a compressed air system in a thermal power plant. Detailed Implementation

[0020] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.

[0021] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0022] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0023] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0024] like Figure 1 As shown in the preferred embodiment of this application, a method for diagnosing the energy consumption of a compressed air system in a thermal power plant includes: S101: Establish a data sensing array for the compressed air system; S102: Acquire operational monitoring data based on the data sensing array, and determine whether to generate an early warning command based on the operational monitoring data and the preset abnormal early warning model; S103: Generate energy consumption diagnosis results based on the early warning instructions and the preset energy consumption diagnosis model; The data sensing array includes multiple monitoring points.

[0025] Specifically, multiple key nodes are selected based on the equipment parameters of the compressed air system, and acquisition sub-modules are set up at each key node to obtain the operating data of the compressed air system.

[0026] Specifically, key nodes include, but are not limited to: equipment and structural points such as the outlets of each air compressor, the areas before and after the dryer, the outlet of the air storage tank, the starting and ending points of branch pipelines in each major functional area, and the final gas consumption point in the pipeline network. Multiple monitoring points are set based on all key nodes, with each monitoring point representing one key node.

[0027] Specifically, a data sensing array is constructed based on all the acquisition sub-modules. The collected operating data includes, but is not limited to: exhaust pressure, exhaust temperature, operating current, loading time, unloading time, and cumulative power consumption of each air compressor; inlet and outlet pressure difference and energy consumption of the dryer; and pipeline pressure and instantaneous flow rate of each key node.

[0028] Specifically, the preset anomaly warning model includes: Acquire historical operating data of the compressed air system and set multiple energy efficiency indicators based on the historical operating data; An initial anomaly warning model was established based on all energy efficiency indicators, and multiple anomaly scenarios were set. Establish a sequence of abnormal scenarios A, A=(a1,a2…ai…an), where ai is the i-th abnormal scenario; n is the number of abnormal scenarios; Based on the abnormal scenario sequence A, ai is sequentially set as the target scenario; Establish a sub-model for recognizing the target scene; Sequentially set up identification sub-models for each abnormal scenario, and construct an abnormal identification model based on all identification sub-models; An anomaly warning model is constructed based on the initial anomaly warning model and the anomaly identification model.

[0029] Specifically, historical operating data refers to the monitoring and recording data of the compressed air system. By filtering and analyzing historical operating data, multiple energy efficiency indicators are set.

[0030] Specifically, the energy efficiency indicators include but are not limited to: the system specific power (the ratio of the total power consumption of all operating air compressors to the total gas production), the specific power of a single air compressor (the ratio of the input power of a single air compressor in the loaded state to the corresponding gas production); the unit pressure drop of the pipe network (the difference between the outlet pressure of the air compressor station and the pressure at the end of each branch pipeline); the total system leakage (the total outlet flow of the air compressor station is directly measured by a flowmeter during non-production hours); the leakage of each branch pipeline (the instantaneous flow at the starting point of each branch pipeline is measured separately during non-production hours); the pressure difference of the dryer (the real-time measured difference between the pressure sensors at the inlet and outlet of the dryer); the load rate of the air compressor (the ratio of the loading time within the statistical period to the sum of the loading time and the unloading time) and other parameters that reflect the energy efficiency state of the compressed air system.

[0031] Specifically, each energy efficiency indicator is quantized so that the reference values of all energy efficiency indicators fall within the same value range.

[0032] Specifically, a plurality of energy consumption fault problems are generated by analyzing historical operating data, and the energy consumption fault problems include but are not limited to: pipe network leakage (a plurality of energy consumption fault problems can be independently set according to different leakage areas), efficiency degradation of the air compressor body (a plurality of energy consumption fault problems can be independently set according to abnormal operating parts of the air compressor), dryer blockage, mismatched operating control parameters, air-using equipment faults and other possible operating fault problems.

[0033] Specifically, a plurality of abnormal scenarios are set according to all the energy consumption fault problems, where a single abnormal scenario represents one energy consumption fault problem.

[0034] Specifically, establishing a recognition sub-model for a target scenario includes: establishing a basic indicator set, where the basic indicator set includes a plurality of basic indicators; generating an associated data packet of the target scenario according to historical operating data; setting the energy efficiency mapping state of the target scenario according to the associated data packet, where the energy efficiency mapping state includes the expected deviation value of each energy efficiency indicator; generating a mapping value of each basic indicator for the target scenario; presetting a mapping value threshold H; if H<hi (i=1, 2...r), setting the i-th basic indicator as a mapping indicator of the target scenario; where hi is the mapping value of the i-th basic indicator for the target scenario; r is the number of basic indicators; acquiring all mapping indicators of the target scenario; setting the basic mapping state of the target scenario according to all the mapping indicators; establishing the recognition sub-model of the target scenario according to the energy efficiency mapping state and the basic mapping state.

[0035] Specifically, relevant operational data matching the target scenario is selected from historical operational data, generating corresponding associated data packages. By analyzing these associated data packages, expected deviation values ​​for each energy efficiency indicator are set (i.e., reference values ​​for each energy efficiency indicator when an energy consumption failure occurs corresponding to the target scenario). Based on all expected deviation values, a corresponding energy efficiency mapping state is constructed.

[0036] Specifically, multiple basic indicators are set based on historical operating data. These basic indicators include, but are not limited to: air compressor exhaust pressure, exhaust temperature, operating current, loading time, unloading time, and cumulative power consumption; dryer inlet and outlet pressure difference and dryer energy consumption; and raw data collected from various monitoring points, such as pipeline pressure and instantaneous flow rate at key nodes. A basic indicator set is constructed based on all these basic indicators.

[0037] Specifically, each basic indicator is quantified so that the reference values ​​of each basic indicator are within the same range.

[0038] Specifically, based on the difference between the current reference value of the basic indicator and the normal reference value (i.e., the reference value when the compressed air system is in normal operation) when the target scenario occurs, a corresponding mapping value is set. The larger the difference (indicating that when the energy consumption fault problem corresponding to the target scenario occurs, the greater the operational fluctuation of the current basic indicator, the stronger the mapping ability for the energy consumption fault problem), the larger the corresponding mapping value. The mapping relationship between the two can be set according to historical parameters.

[0039] Specifically, the mapping threshold can be set according to historical parameters. If the current basic indicator's mapping value is greater than the preset mapping threshold, it means that the current basic indicator can map the energy consumption fault problem corresponding to the target scenario. It is then set as the mapping indicator for the target scenario, and the corresponding mapping value is set according to historical operating parameters (i.e., the reference value corresponding to the basic indicator when the energy consumption fault problem corresponding to the target scenario occurs). The basic mapping state is constructed based on all mapping values.

[0040] It is understood that, in the above embodiments, by constructing a dual diagnostic structure of energy efficiency mapping state and basic mapping state, the potential energy consumption fault range can be accurately identified, the rapid diagnosis of energy consumption anomalies can be improved, and the energy consumption diagnosis efficiency of the compressed air system of thermal power plants can be improved.

[0041] In a preferred embodiment of this application, determining whether to generate a warning instruction includes: Multiple feedback time points can be preset; Acquire the operational monitoring data obtained by the data sensing array at the current monitoring time point; Generate an energy efficiency anomaly value f based on operational monitoring data; f=[ si]; wherein θ1 is the number of energy efficiency indicators; si is the deviation value of the i-th energy efficiency indicator generated according to the operation monitoring data; preset energy efficiency abnormality value threshold F1; If f>F1, an early warning instruction is generated at the current feedback time node.

[0042] Specifically, the early warning instruction includes: establishing an initial operation state according to the deviation values of the respective energy efficiency indicators; generating primary similarity values between the initial operation state and the respective abnormal scenarios; preset primary similarity value threshold K1; If K1<ki, (i=1, 2...n), the i-th abnormal scenario is set as a to-be-judged scenario; wherein ki is the primary similarity value between the initial operation state and the i-th abnormal scenario; n is the number of abnormal scenarios; setting an auxiliary collection instruction according to all the to-be-judged scenarios, and acquiring an auxiliary data packet according to the auxiliary collection instruction.

[0043] Specifically, real-time reference values of the respective energy efficiency indicators are generated according to the operation monitoring data, and the corresponding deviation values are set according to the difference between the real-time reference values and standard reference values (i.e., the reference values of the respective energy efficiency indicators when the compressed air system is in a normal operation state). The larger the difference, the larger the corresponding deviation value, and the mapping relationship between the two can be set according to historical parameters.

[0044] Specifically, the energy efficiency abnormality value threshold can be set according to historical parameters. When the real-time energy efficiency abnormality value is greater than the preset energy efficiency abnormality value threshold, it indicates that the current compressed air system has a potential risk of energy efficiency failure, and further diagnosis needs to be performed according to the early warning instruction.

[0045] Specifically, the corresponding primary similarity value is set according to the degree of difference between the initial operation state and the reference values of the respective energy efficiency indicators corresponding to the current abnormal scenario. The smaller the degree of difference, the larger the corresponding primary similarity value, and the mapping relationship between the two can be set according to historical parameters.

[0046] Specifically, the primary similarity value threshold can be set according to historical parameters. If the primary similarity value of the current abnormal scenario is greater than the preset primary similarity value threshold, it indicates that the energy consumption failure problem corresponding to this abnormal scenario may exist. It is necessary to acquire the real-time parameters of each mapping indicator in the basic mapping state corresponding to the abnormal scenario according to the auxiliary collection instruction, generate the corresponding auxiliary sub-packet, and generate the auxiliary data packet according to the collected auxiliary sub-packets corresponding to the respective to-be-judged scenarios.

[0047] In the preferred embodiment of the examples of the present application, according to the preset energy consumption diagnosis model, the steps include: Based on the abnormal scenario sequence A, ai are sequentially set as the scenarios to be diagnosed; Define a diagnostic sub-strategy for the scenario to be diagnosed, which includes multiple judgment components. Configure diagnostic sub-strategies for each abnormal scenario in sequence; An energy consumption diagnostic model is constructed based on all diagnostic sub-strategies.

[0048] Specifically, by analyzing historical operating data, multiple judgment sub-components are set for each scenario to be diagnosed (i.e., for the energy efficiency fault problem corresponding to the scenario to be diagnosed, a corresponding diagnostic process is set, and the diagnostic process is divided into multiple sub-processes, with each sub-process serving as a judgment sub-component. For example, when judging the energy consumption fault problem of reduced efficiency of the air compressor, the actual specific power of a single air compressor under load is compared with the rated specific power or the historical best specific power of the same model of air compressor. If the actual specific power is consistently higher than the benchmark value by 10%, and at the same time, it is checked whether the exhaust temperature of the air compressor is more than 5 degrees Celsius higher than the normal value and whether the load current is more than 5% higher than the normal value, then it is determined that the efficiency of the air compressor has decreased. The exhaust temperature judgment process, the actual power judgment process, and the load current judgment process are set as judgment sub-components). Specifically, relevant data is collected based on the diagnostic sub-strategy, and the collected data is analyzed to determine whether there are corresponding energy consumption faults, thereby generating the final energy consumption diagnostic report.

[0049] Specifically, generating energy consumption diagnostic results includes: Generate secondary similarity values ​​with each scenario to be judged based on all auxiliary data packets; The primary diagnostic order is determined based on all secondary similarity values; Execution scenarios are selected sequentially based on the order of primary diagnostics; Diagnostic feedback data is obtained based on the diagnostic sub-strategy of the execution scenario, and a matching value w is generated based on the diagnostic feedback data; Preset matching threshold W1; If w > W1, generate a first-level diagnostic instruction for the execution scenario; Retrieve all Level 1 diagnostic commands and generate diagnostic results.

[0050] Specifically, a secondary reference value is set based on the difference between the real-time reference value of each mapping indicator in the auxiliary sub-package of the current scene to be judged and the reference value of each mapping indicator in the basic mapping state corresponding to the current scene to be judged. The smaller the difference, the larger the corresponding secondary similarity value. The mapping relationship between the two can be set according to historical parameters.

[0051] Specifically, the larger the secondary similarity value corresponding to the current scenario to be judged, the greater the possibility of a corresponding energy efficiency fault, and the higher its position in the primary diagnostic sequence. By dynamically adjusting the diagnostic sequence, the efficiency of energy consumption diagnosis of the compressed air system of thermal power plants can be improved.

[0052] Specifically, relevant data is collected based on the diagnostic sub-strategy, generating a matching degree between the collected data and each judgment sub-component in the execution scenario. A corresponding fit value is set based on the total matching degree; the higher the total matching degree, the larger the corresponding fit value. The mapping relationship between the two can be set based on historical parameters. When the fit value of the execution scenario exceeds a preset fit value threshold, a first-level diagnostic instruction is generated. This first-level diagnostic instruction indicates that the compressed air system has an energy consumption fault corresponding to the execution scenario at the current time point. By summarizing all first-level diagnostic instructions, the final energy consumption diagnostic result is generated.

[0053] It is understood that in the above embodiments, the abnormal early warning model performs an initial screening of energy consumption problems based on the real-time energy consumption deviation status, and dynamically adjusts the diagnostic strategy based on the screening results and the energy consumption diagnosis model to quickly diagnose abnormal energy consumption problems and improve the energy consumption diagnosis efficiency of the compressed air system of thermal power plants.

[0054] In another preferred embodiment of the energy consumption diagnosis method for compressed air systems in thermal power plants based on any of the above preferred embodiments, this preferred embodiment provides an energy consumption diagnosis system for compressed air systems in thermal power plants, comprising: The monitoring unit is used to establish a data sensing array for the compressed air system; The data sensing array includes multiple monitoring points; The monitoring unit includes multiple data acquisition sub-modules, which are set up at various monitoring points; The central control unit acquires operational monitoring data through a data sensing array; The central control unit includes: The first processing module is used to establish an anomaly early warning model; The first processing module is also used to determine whether to generate an early warning instruction based on the operation monitoring data and the anomaly early warning model; The second processing module is used to establish an energy consumption diagnostic model; The second processing module is also used to generate energy consumption diagnosis results based on the early warning instructions and the energy consumption diagnosis model.

[0055] Specifically, the acquisition submodule is preferably a data acquisition device of various types, which can be set according to the data types to be acquired at each monitoring point.

[0056] In a preferred embodiment of this application, the first processing module is further configured to: Acquire historical operation data of the compressed air system, and set a plurality of energy efficiency indicators according to the historical operation data; Establish an initial abnormal early warning model according to all the energy efficiency indicators, and set a plurality of abnormal scenarios; Establish an abnormal scenario sequence A, where A=(a1, a2...ai...an), ai is the i-th abnormal scenario, and n is the number of abnormal scenarios; Set ai as the target scenario in sequence according to the abnormal scenario sequence A; Establish a recognition sub-model of the target scenario; Set the recognition sub-model of each abnormal scenario in sequence, and construct an anomaly recognition model according to all the recognition sub-models; Construct an abnormal early warning model according to the initial abnormal early warning model and the anomaly recognition model; Wherein, establishing the recognition sub-model of the target scenario includes: Establish a basic indicator set, where the basic indicator set includes a plurality of basic indicators; Generate an associated data packet of the target scenario according to the historical operation data; Set the energy efficiency mapping state of the target scenario according to the associated data packet, where the energy efficiency mapping state includes the expected deviation value of each energy efficiency indicator; Generate the mapping value of each basic indicator to the target scenario; Preset a mapping value threshold H; If H<hi (i=1, 2...r), set the i-th basic indicator as the mapping indicator of the target scenario; Wherein, hi is the mapping value of the i-th basic indicator to the target scenario, and r is the number of basic indicators; Acquire all mapping indicators of the target scenario; Set the basic mapping state of the target scenario according to all the mapping indicators; Establish the recognition sub-model of the target scenario according to the energy efficiency mapping state and the basic mapping state.

[0057] In the preferred embodiment of the examples of the present application, the first processing module is further configured to: Preset a plurality of feedback time nodes; Acquire operation monitoring data acquired by the data sensing array at the current monitoring time node; Generate an energy efficiency abnormal value f according to the operation monitoring data; f= si]; Wherein, θ1 is the number of energy efficiency indicators, and si is the deviation value of the i-th energy efficiency indicator generated according to the operation monitoring data; Preset an energy efficiency abnormal value threshold F1; If f>F1, generate an early warning instruction at the current feedback time node; wherein the warning instruction includes: establishing an initial operating state according to the deviation value of each energy efficiency indicator; generating primary similarity values between the initial operating state and each abnormal scenario; presetting a primary similarity value threshold K1; if K1<ki (i=1, 2...n), setting the i-th abnormal scenario as a to-be-judged scenario; wherein ki is the primary similarity value between the initial operating state and the i-th abnormal scenario; n is the number of abnormal scenarios; setting an auxiliary acquisition instruction according to all the to-be-judged scenarios, and acquiring an auxiliary data packet according to the auxiliary acquisition instruction.

[0058] According to the first concept of the present application, by adding a data sensing array to collect the operating data of the compressed air system, the energy consumption deviation state of the compressed air system is monitored in real time, and the dual diagnosis mechanism of the abnormal warning model and the energy consumption diagnosis model is added to improve the diagnosis accuracy of the compressed air system in the thermal power plant.

[0059] According to the second concept of the present application, the abnormal warning model performs initial screening on energy consumption problems according to the real-time energy consumption deviation state, and dynamically adjusts the diagnosis strategy in combination with the energy consumption diagnosis model according to the screening result, so as to quickly diagnose the energy consumption abnormal problem and improve the energy consumption diagnosis efficiency of the compressed air system in the thermal power plant.

[0060] The above description is merely the preferred embodiment of the present application. It should be noted that for those of ordinary skill in the art, several improvements and substitutions may be made without departing from the technical principle of the present application, and these improvements and substitutions shall also be regarded as falling within the protection scope of the present application.

Claims

1. A method for diagnosing energy consumption in a compressed air system of a thermal power plant, characterized in that, comprising: establishing a data sensing array for the compressed air system; acquiring operation monitoring data based on the data sensing array, and judging whether to generate an early warning instruction according to the operation monitoring data and a preset abnormal early warning model; generating an energy consumption diagnosis result according to the early warning instruction and a preset energy consumption diagnosis model; wherein the data sensing array comprises a plurality of monitoring points.

2. The energy consumption diagnosis method for compressed air systems in thermal power plants as described in claim 1, characterized in that, The preset abnormal early warning model comprises: acquiring historical operation data of the compressed air system, and setting a plurality of energy efficiency indexes according to the historical operation data; establishing an initial abnormal early warning model based on all the energy efficiency indexes, and setting a plurality of abnormal scenarios; establishing an abnormal scenario sequence A, where A=(a1,a2...ai...an), where ai is the i-th abnormal scenario, and n is the number of abnormal scenarios; sequentially setting ai as a target scenario according to the abnormal scenario sequence A; establishing a recognition sub-model for the target scenario; sequentially setting the recognition sub-model for each abnormal scenario, and constructing an abnormal recognition model according to all the recognition sub-models; constructing the abnormal early warning model according to the initial abnormal early warning model and the abnormal recognition model.

3. The energy consumption diagnosis method for compressed air systems in thermal power plants as described in claim 2, characterized in that, The establishing the recognition sub-model for the target scenario comprises: establishing a basic index set, wherein the basic index set comprises a plurality of basic indexes; generating an associated data packet of the target scenario according to the historical operation data; setting an energy efficiency mapping state of the target scenario according to the associated data packet, wherein the energy efficiency mapping state comprises an expected deviation value of each energy efficiency index; generating a mapping value of each basic index to the target scenario; presetting a mapping value threshold H; if H<hi (i=1, 2...r), setting the i-th basic index as a mapping index of the target scenario; wherein hi is the mapping value of the i-th basic index to the target scenario; r is the number of basic indexes; acquiring all mapping indexes of the target scenario; setting a basic mapping state of the target scenario according to all the mapping indexes; establishing the recognition sub-model of the target scenario according to the energy efficiency mapping state and the basic mapping state.

4. The energy consumption diagnosis method for compressed air systems in thermal power plants as described in claim 3, characterized in that, The judging whether to generate an early warning instruction comprises: presetting a plurality of feedback time nodes; acquiring the operation monitoring data acquired by the data sensing array at the current monitoring time node; generating an energy efficiency abnormal value f according to the operation monitoring data; f=[ you]; wherein θ1 is the number of energy efficiency indexes; si is the deviation value of the i-th energy efficiency index generated according to the operation monitoring data; presetting an energy efficiency abnormal value threshold F1; if f>F1, generating the early warning instruction at the current feedback time node.

5. The energy consumption diagnosis method for compressed air systems in thermal power plants as described in claim 4, characterized in that, The early warning instruction comprises: establishing an initial operation state according to the deviation value of each energy efficiency index; generating a first-level similarity value between the initial operation state and each abnormal scenario; presetting a first-level similarity value threshold K1; if K1<ki, (i=1, 2...n), setting the i-th abnormal scenario as a to-be-judged scenario; wherein ki is the first-level similarity value between the initial operation state and the i-th abnormal scenario; n is the number of abnormal scenarios; setting an auxiliary acquisition instruction according to all the to-be-judged scenarios, and acquiring an auxiliary data packet according to the auxiliary acquisition instruction.

6. The energy consumption diagnosis method for compressed air systems in thermal power plants as described in claim 5, characterized in that, The preset energy consumption diagnosis model comprises: sequentially setting ai as a to-be-diagnosed scenario according to the abnormal scenario sequence A; setting a diagnosis sub-strategy for the to-be-diagnosed scenario, wherein the diagnosis sub-strategy comprises a plurality of judgment sub-components; sequentially setting the diagnosis sub-strategy for each abnormal scenario; Construct an energy consumption diagnosis model based on all diagnosis sub-strategies.

7. The energy consumption diagnosis method for compressed air systems in thermal power plants as described in claim 6, characterized in that, The generating the energy consumption diagnosis result includes: Generating secondary similarity values corresponding to each to-be-judged scenario based on all auxiliary data packets; Setting a primary diagnosis order according to all the secondary similarity values; Selecting execution scenarios sequentially based on the primary diagnosis order; Acquiring diagnosis feedback data according to the diagnosis sub-strategy of the execution scenario, and generating a fitness value w based on the diagnosis feedback data; Presetting a fitness value threshold W1; If w>W1, generating a primary diagnosis instruction for the execution scenario; Acquiring all primary diagnosis instructions, and generating a diagnosis result.

8. A compressed air system energy consumption diagnosis system for thermal power plants, employing the energy consumption diagnosis method for compressed air systems of thermal power plants according to any one of claims 1-7, characterized in that, Comprising: A monitoring unit, configured to establish a data sensing array for a compressed air system; A plurality of monitoring points are included in the data sensing array; The monitoring unit comprises a plurality of collection submodules, and the collection submodules are arranged at each monitoring point; A central control unit, configured to acquire operation monitoring data through the data sensing array; The central control unit comprises: A first processing module, configured to establish an abnormality early warning model; The first processing module is further configured to judge whether to generate an early warning instruction according to the operation monitoring data and the abnormality early warning model; A second processing module, configured to establish an energy consumption diagnosis model; The second processing module is further configured to generate an energy consumption diagnosis result according to the early warning instruction and the energy consumption diagnosis model.

9. The energy consumption diagnostic system for compressed air systems in thermal power plants as described in claim 8, characterized in that, The first processing module is further configured to: Acquire historical operation data of the compressed air system, and set a plurality of energy efficiency indicators according to the historical operation data; Establish an initial abnormality early warning model based on all the energy efficiency indicators, and set a plurality of abnormality scenarios; Establish an abnormality scenario sequence A, where A=(a1,a2…ai…an), ai is the i-th abnormality scenario, and n is the number of abnormality scenarios; Sequentially set ai as the target scenario according to the abnormality scenario sequence A; Establish a recognition sub-model for the target scenario; Sequentially set the recognition sub-model for each abnormality scenario, and construct an abnormality recognition model based on all the recognition sub-models; Construct the abnormality early warning model based on the initial abnormality early warning model and the abnormality recognition model; The establishing the recognition sub-model for the target scenario comprises: Establish a basic indicator set, where a plurality of basic indicators are included in the basic indicator set; Generate an associated data packet of the target scenario according to the historical operation data; Set the energy efficiency mapping state of the target scenario according to the associated data packet, where the energy efficiency mapping state includes the expected deviation value of each energy efficiency indicator; Generate a mapping value of each basic indicator for the target scenario; Preset a mapping value threshold H; If H<hi (i=1,2…r), set the i-th basic indicator as the mapping indicator of the target scenario; Where hi is the mapping value of the i-th basic indicator for the target scenario; r is the number of basic indicators; Acquire all mapping indicators of the target scenario; Set the basic mapping state of the target scenario according to all the mapping indicators; Establish the recognition sub-model of the target scenario according to the energy efficiency mapping state and the basic mapping state.

10. The energy consumption diagnostic system for compressed air systems in thermal power plants as described in claim 9, characterized in that, The first processing module is further configured to: Preset a plurality of feedback time nodes; Acquire the operation monitoring data acquired by the data sensing array at the current monitoring time node; Generate an energy efficiency abnormality value f according to the operation monitoring data; f=[ you]; Where θ1 is the number of energy efficiency indicators; si is the deviation value of the i-th energy efficiency indicator generated according to the operation monitoring data; Preset an energy efficiency abnormality value threshold F1; If f>F1, an early warning instruction is generated at the current feedback time node; wherein, the early warning instruction comprises: establishing an initial operation state according to the deviation value of each energy efficiency index; generating primary similarity values between the initial operation state and each abnormal scenario; presetting a primary similarity value threshold K1; If K1<ki, (i=1, 2...n), setting the i-th abnormal scenario as a to-be-judged scenario; wherein, ki is the primary similarity value between the initial operation state and the i-th abnormal scenario; n is the number of abnormal scenarios; setting an auxiliary acquisition instruction according to all the to-be-judged scenarios, and acquiring an auxiliary data packet according to the auxiliary acquisition instruction.

Images