A fluid measurement and control device whole life cycle operation and maintenance management system based on multi-source data

Abstract

The application relates to the technical field of equipment asset management and operation and maintenance decision-making, and discloses a fluid measurement and control equipment whole life cycle operation and maintenance management system based on multi-source data, which comprises the following steps: allocating an initial quota of operation and maintenance depreciation rights to equipment; acquiring multi-source data reflecting physical loss, business value, real-time operation efficiency and operation stability of the equipment; dynamically calculating the consumption rate of ODR based on the product of a plurality of factors determined based on the multi-source data; and driving operation and maintenance decision-making according to the remaining quota. The application converts the physical loss process of the equipment into a value budget management process in which the consumption cost is adjusted in real time according to the business value, operation efficiency and stability risk, establishes a quantitative bridge connecting the physical world and business decision-making, enables operation and maintenance decision-making to be based on clear economic basis instead of a single physical threshold, and thus improves the asset management level and decision-making quality.

Description

TECHNICAL FIELD

[0001] The present application relates to a kind of fluid measurement and control equipment full life cycle operation and maintenance management system based on multi-source data, belong to equipment asset management and operation and maintenance decision-making technical field. BACKGROUND

[0002] At present, especially in the full life cycle management of fluid measurement and control equipment, the management system based on condition monitoring or predictive maintenance is generally used, these systems obtain various physical parameters, such as vibration, temperature, pressure data, when the equipment is running, to evaluate the real-time health status of the equipment, and trigger maintenance warning or intervention action when the index exceeds the preset threshold, compared with traditional periodic maintenance or post-failure maintenance, which improves the reliability and availability of the equipment.

[0003] However, when such management system is applied to high-value continuous industrial processes, one of the design assumptions on which it depends, that is, the physical wear and tear of the equipment is an objective and homogeneous process, has limitations under certain conditions, the operation basis of the existing technology system is that the management weight of a unit of physical wear and tear, for example, a 1% decrease in equipment health index, is equal at any time and under any business background, but in actual production and operation, the actual economic impact of a small physical wear and tear is highly related to the business context in which it occurs, a physical wear and tear occurring in daily smooth production is different from a physical wear and tear of the same degree occurring before the delivery of a high-value order or a major maintenance window, the latter poses a potential economic risk to the entire production plan.

[0004] To deal with this problem, some intuitive improvement paths, such as simply improving monitoring accuracy to capture earlier physical degradation, or attaching static business importance labels to devices, cannot solve the problem. The former will interfere with normal production rhythm due to a large number of physically correct but business insignificant alarms, and the latter cannot dynamically quantify the loss cost of the device in different business value periods due to its static and qualitative properties. Managers still cannot clearly weigh the real economic cost between the two decisions of continuing to run to complete the current task and shutting down to protect the long-term value of the device. Specifically, the prior art mainly has the following deficiencies: 1. Lack of a running mechanism that quantitatively associates device physical wear and tear with dynamically changing business value, resulting in a lack of effective quantitative association between information required for operation and maintenance decisions and enterprise business goal information; 2. The evaluation of the health status of the device stays in the physical dimension and cannot measure its value state in a specific business period, so that risk management cannot cover the economic risks caused by changes in business context. Therefore, how to build an operation and maintenance management system that not only monitors the physical health of the device, but also establishes an asset value accounting and budget management mechanism that dynamically accounts for the correlation between physical wear and tear and business context, so that operation and maintenance decisions can be based on clear quantitative economic value rather than a single physical threshold, has become a technical problem to be solved by the present application. SUMMARY

[0005] The present application provides a kind of fluid measurement and control equipment full life cycle operation and maintenance management system based on multi-source data, its main purpose is to solve the problem in prior art, equipment physical wear and tear and its economic value of the business process it is in are disconnected, resulting in operation and maintenance decision lacks quantitative economic basis.

[0006] To achieve the above purpose, the present application provides a kind of fluid measurement and control equipment full life cycle operation and maintenance management system based on multi-source data, the system includes:

[0007] An operation and maintenance loss right ODR quota allocation module is configured to allocate an initial ODR total quota to a fluid measurement and control device at the beginning of a management period.

[0008] An ODR consumption dynamic accounting module is configured to obtain operation data reflecting physical loss of the device to determine a base physical loss rate, obtain management data reflecting business process context of the device to determine a business context multiplier, obtain real-time actual power consumption and theoretical power consumption of the device to determine a process efficiency correction factor, and obtain high-frequency state data of the device and calculate information entropy of the high-frequency state data within a time window to determine a state fluctuation risk multiplier; the ODR consumption dynamic accounting module is further configured to calculate a real-time ODR consumption rate of the device based on a product of the base physical loss rate, the business context multiplier, the process efficiency correction factor, and the state fluctuation risk multiplier.

[0009] An ODR budget management module is configured to continuously update a remaining ODR quota of the device according to the initial ODR total quota and the real-time ODR consumption rate.

[0010] A decision driving module is configured to trigger one or more predefined operation and maintenance management actions when the remaining ODR quota meets one or more pre-set decision conditions.

[0011] Preferably, the system further comprises an ODR supplement module configured to determine an ODR supplement corresponding to a maintenance operation according to a type of the maintenance operation and a predetermined effect verification procedure, and use the ODR supplement to increase the remaining ODR quota of the device after the maintenance operation is performed on the fluid measurement and control device.

[0012] Preferably, the ODR consumption dynamic accounting module is further configured to obtain a real-time electric energy meter reading of the device as the actual power consumption, and establish a benchmark model mapping a current working condition to the theoretical power consumption based on design performance curve data or historical optimal working condition data of the device, and determine the theoretical power consumption by the benchmark model; wherein the process efficiency correction factor is determined in the following manner: wherein is the actual power consumption, is the theoretical power consumption.

[0013] Preferably, the determination of the state fluctuation risk multiplier in the ODR consumption dynamic accounting module is achieved in the following manner: obtaining a data sequence of the high-frequency state data from a vibration sensor or an acoustic sensor of the fluid measurement and control device; using a sample entropy algorithm to calculate complexity of the data sequence within a time window, and normalizing the calculation result to generate a scalar value quantifying uncertainty of a running state of the device as the state fluctuation risk multiplier; wherein a length of the time window and a calculation step are pre-set.

[0014] Preferably, the system further comprises a fluctuation pattern analysis module configured to: perform a fast Fourier transform on the data sequence of the high-frequency state data in parallel to extract a frequency energy distribution feature thereof; perform a pattern matching calculation on the frequency energy distribution feature against a preset prototype library storing a plurality of typical fluctuation prototypes; and the decision driving module is further configured to output a classification annotation of the matched typical fluctuation prototype as decision reference information when triggering the operation and maintenance management action.

[0015] Preferably, the ODR consumption dynamic accounting module is further configured to: determine the value of the business context multiplier according to a preset rule mapping the management data to the business context multiplier; and the management data comprises production plan information, order value information, and criticality level information of the fluid measurement and control device in the process flow.

[0016] Preferably, the decision driving module is further configured to provide a human-computer interaction interface responsive to a simulation request input by a user and containing a future business scenario and a running time, and to call the ODR consumption dynamic accounting module to predict a future consumption trajectory of the device remaining ODR quota under the simulation request.

[0017] Preferably, the system further comprises an asset behavior portrait analysis module configured to: extract a set of management feature indicators representing economic behavior of the device based on ODR consumption history data and ODR replenishment history data of the fluid measurement and control device in one or more historical management cycles; and determine a behavior portrait classification for each device by using an unsupervised clustering algorithm according to the management feature indicators; wherein the ODR quota allocation module determines an initial total ODR quota allocated to the device in the next management cycle according to the behavior portrait classification of the device by applying a set of differentiated allocation rules.

[0018] Preferably, the system further comprises: an asset relationship definition module configured to determine that there is a functional substitutability relationship between the first fluid measurement and control device and the second fluid measurement and control device; and an ODR budget accommodation module configured to, when a fusion condition that the remaining ODR quota of the first fluid measurement and control device is lower than a first preset threshold and the remaining ODR quota of the second fluid measurement and control device is higher than a second preset threshold is met, transfer part of the ODR quota of the second fluid measurement and control device to the ODR quota of the first fluid measurement and control device according to an accommodation cost coefficient preset in a range of 1.05 to 1.3.

[0019] Preferably, the system further comprises: a strategic ODR issuance module configured to, in response to an authorization instruction, add a temporary ODR credit quota to the fluid measurement and control device when a strategic business event meeting a condition is identified; and a maintenance liability management module configured to, while the ODR credit quota is added, generate a future maintenance liability associated with the credit quota, and trigger a compensatory maintenance work order based on the future maintenance liability in a future business window.

[0020] Compared with the prior art, the present application has the beneficial effects that:

[0021] 1. By means of the operation and maintenance loss quota allocation module, the collaborative work of the dynamic accounting module and the budget management module is consumed, and an asset operation and maintenance mode that binds the device physical loss and the business process context is established. The system allocates an ODR total quota to the device at the beginning of the management cycle, which defines the total amount of loss that can be tolerated by the device in the cycle. During operation, the ODR consumption dynamic accounting module obtains device physical loss data, and combines the price multiplier determined by the management data reflecting the value of the business process to continuously calculate the consumption rate of ODR. Finally, the ODR budget management module updates the quota balance, and the decision-driven module triggers the operation and maintenance action according to the quota state. This operation mode changes the physical loss process of the device from an objective physical process determined only by time or intensity into a management process whose consumption cost is adjusted by the business value in real time, providing a judgment basis that directly reflects the business economy independent of the physical state for operation and maintenance decision-making.

[0022] 2. The ODR consumption dynamic accounting module also obtains process efficiency data reflecting the real-time operation efficiency of the device, and obtains high-frequency state data reflecting the stability of the operation state of the device, respectively determines the process efficiency correction factor and the state fluctuation risk multiplier, and corrects the ODR consumption rate based on the same. This mechanism integrates the management data, operation efficiency data and state stability data into the ODR consumption accounting model, so that the ODR cost paid for a single physical loss takes into account the energy conversion efficiency of the business value generated by the loss at the time of the loss and the operation stability risk of the device at that time, which avoids the situation of equivalent processing of the same physical loss under different efficiencies and different risks, so that the consumption of ODR budget more truly reflects the comprehensive loss currently borne by the device.

[0023] 3. This system further incorporates an asset behavior profile analysis module and an ODR budget allocation module. The former, based on the ODR consumption and replenishment data of devices within historical management cycles, defines a behavior profile category for each device. The ODR quota allocation module will then allocate quotas differentiatedly based on this category in the next management cycle. The latter, among devices with pre-defined functional substitutability, allows for internal transfer of quotas based on the condition that one device has a low ODR quota while another has a sufficient quota. The introduction of these two modules expands the ODR management system from independent management focused on a single device and a single cycle to adaptive resource allocation across cycles for device clusters. As a result, the system possesses the ability to self-optimize budget allocation strategies based on historical economic behavior data and dynamically balance and adjust ODR resources within the system, enhancing the resource allocation flexibility of the entire device group when dealing with localized high-intensity business impacts. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the core functional modules and closed-loop data flow of the system of the present invention;

[0025] Figure 2 This is a schematic diagram of the multi-layered deployment architecture required for the implementation of the system of this invention;

[0026] Figure 3 This is an interactive timing diagram of the dynamic adjustment of the device health quota in this invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] The present invention provides a full lifecycle operation and maintenance management system for fluid measurement and control equipment based on multi-source data. Its overall architecture mainly includes an operation and maintenance loss right. Quota allocation module, one Consumption dynamic accounting module, one The module includes a budget management module and a decision-driven module; The quota allocation module is responsible for setting an initial quota for the fluid measurement and control equipment managed by the system at the beginning of a management cycle. The total quota, which is then handed over to The budget management module tracks and updates the entire lifecycle of the equipment; during equipment operation, The dynamic consumption calculation module continuously obtains information from multiple data sources and dynamically calculates the device's real-time consumption. Consumption rate, and pass that rate data to The budget management module is used to deduct the remaining credit limit; ultimately, the decision-driven module bases its decisions on... The budget management module provides the credit limit status, which triggers corresponding operation and maintenance management actions when preset conditions are met, thus forming a closed-loop management process.

[0029] In a specific application scenario, such as the core process management of a large-scale refining and chemical enterprise, the safe and stable operation of fluid monitoring and control equipment is directly related to the economic benefits of the entire production unit. A common challenge of existing operation and maintenance management methods is that the assessment of equipment physical wear and tear is disconnected from the business environment in which it occurs, resulting in a lack of clear economic quantification for management decisions. To address this challenge, the system of this invention sets an Operational Depreciation Right. The quota allocation module transforms the physical health of equipment from a passively monitored status indicator into a proactively manageable, economically-oriented, and consumable budget. The quota allocation module allocates an initial quota to a fluid monitoring and control device at the beginning of a management cycle, such as the start of a fiscal year or a production quarter. The total amount is determined as follows: First, input the design life characteristics of the equipment, such as the mean time between failures (MTBF) data provided by the equipment manufacturer, and combine it with the historical performance degradation curve data of the equipment within the enterprise; Second, input the enterprise's core business objectives for this management cycle, such as the annual total output target or the overall equipment efficiency (OEE) target. The quota allocation module has a built-in benchmark model that transforms the above input into a standardized dimensionless value. Total amount, for example, setting an initial annual quota of 10,000 standard loss units for the feedstock supply pump group of a core cracking unit ( Thus, an uncertain and continuous physical decay process is transformed into a budget management problem with a fixed total amount and auditability, laying the foundation for subsequent value-based operation and maintenance management; furthermore, to address the issue that the economic impact of physical losses is highly correlated with the business context in which they occur, the system is configured with... The dynamic accounting module consumes resources. Its core task is to dynamically calculate the product of multiple factors determined from multi-source data. The consumption rate; its complete calculation logic is determined as follows: ,in, Based on the basic physical loss rate For business context multipliers, This is a process efficiency correction factor. This is the risk multiplier for state fluctuations.

[0030] The determination of the basic physical wear rate (BDR) involves acquiring operational data reflecting the physical wear of the equipment, such as cumulative operating hours, rotational speed, or stroke count. A basic BDR rate is then derived based on a static lookup table or empirical formula derived from historical equipment data. The procedure for establishing the lookup table or model used to determine the BDR for this fluid control equipment includes the following steps: First, retrieve historical operating data and corresponding maintenance records for at least one complete overhaul cycle of the equipment or similar equipment. These maintenance records contain status assessments or replacement information for core vulnerable components such as bearings and seals. Second, define each core component replacement event that restores equipment performance to more than 95% of the factory design standard as the zero-point anchor for physical wear. The cumulative operating parameters between two anchor points, such as cumulative operating hours or equivalent revolutions, are used as independent variables for loss evolution. Next, based on the wear quantification data or performance degradation curves of vulnerable parts under historical replacement, a Weibull distribution or exponential function is used to nonlinearly fit the relationship between the loss degree and the independent variables, resulting in a loss accumulation function. Finally, the first derivative of this loss accumulation function is calculated to obtain the loss rate function, which is then discretized according to the cumulative operating parameters to generate a lookup table with the cumulative operating parameters as input and the basic physical loss rate as output. 100% of the total loss degree is calibrated as corresponding to an initial total ODR (Operational Defect Rate). For example, for the aforementioned pump set, each hour of stable operation under rated conditions is determined by the model to generate [loss rate]. indivual The basic loss, i.e. , The determination of this is intended to provide a real-time valuation of the business value of physical losses. The dynamic consumption accounting module is configured to acquire management data reflecting the context of equipment business processes. This management data is obtained through standardized data interfaces with Enterprise Resource Planning (ERP) or Manufacturing Execution System (MES). The interaction path prioritizes encrypted data exchange via the enterprise's intranet. In backup cases of network outages, authorized personnel can import the data offline using encrypted storage media. The management data includes order value information associated with the production plan information of the current production batch, or the criticality level information of the equipment in the current process flow. The module is internally configured with a mapping rule engine that can be set by administrators to map different management data states to specific... Numerical values, for example, when executing a regular production plan, Set as During the sprint production phase to ensure a high-value export order, the rules engine automatically... Upgraded to This makes the same physical wear and tear generated during critical business periods consume ten times more health budget.

[0031] The purpose of the determination is to modify the accounting of physical wear and tear itself to avoid equating the same physical wear and tear generated by high-efficiency operation and low-efficiency idling; to achieve this purpose, the consumption dynamic accounting module is configured to obtain the real-time actual power consumption (Actual Power, ) and theoretical power consumption (Theoretical Power, ) of the device to determine the process efficiency correction factor ; wherein, The acquisition is realized by reading the data of the digital electric energy meter in the power distribution cabinet of the device through the standard industrial bus protocol (such as Modbus) in real time, which utilizes the existing device without adding new hardware; The determination is made at the initial stage of system deployment through an offline modeling procedure, which utilizes the factory performance curve data or historical optimal operating condition flow-pressure-power correspondence data of the device to establish a baseline model that maps the current operating condition (such as real-time flow, pressure) to the theoretical power consumption; The determination method is ; when the device runs inefficiently due to cavitation or abnormal medium viscosity, etc., its will be higher than under the same operating condition, for example while , then , which makes the consumption rate of suffer a punitive addition, thus reflecting the additional physical stress the device bears due to inefficient operation.

[0032] The determination aims to introduce a risk premium due to uncertainty itself into the pricing system of to quantify the risk of severe fluctuations in the state of the device before critical failure; for this purpose, the consumption dynamic accounting module is also configured to obtain high-frequency state data of the device and calculate the information entropy of the high-frequency state data within a time window to determine the state fluctuation risk multiplier ; the calculation of the state fluctuation risk multiplier CVE is based on a band-limited linear gain function calibrated using historical data, and the specific procedure is as follows: first, select the high-frequency state data segments of the device under multiple verified, fault-free stable operating conditions from the historical data set, calculate the sample entropy of these data segments, and determine the statistical average of the obtained sample entropy as the baseline entropy value ; second step, the real-time value of CVE is determined by the following function: , wherein, is the sample entropy value calculated in real time within the current time window, is an upper limit value of the risk multiplier set according to the criticality of the equipment and the risk tolerance, for example, 5.0, and the sensitivity coefficient is determined by replaying a complete historical fault evolution data containing from normal to near failure, adjusting until the change curve of CVE can meet the preset warning time window while avoiding frequent false alarms caused by instantaneous working condition fluctuations; in specific implementation, the system uses the existing vibration sensor or acoustic sensor of the equipment to obtain a data sequence of high-frequency state data at a sampling rate of no less than ; and uses the sample entropy (Sample Entropy) algorithm to calculate the complexity of the data sequence within a sliding time window with a length of seconds and a calculation step of seconds, and the calculation result is normalized as a scalar value as ; when the equipment is running smoothly, the signal is regular, and the calculated value is close to , and when the equipment produces irregular impact signals due to the impending rupture of the bearing, the signal complexity increases sharply, which may cause the value to jump to or higher, so that its consumption rate is multiplied by a risk coefficient.

[0033] Taking the aforementioned pump set as an example, a complete numerical deduction is carried out: at a certain moment, the pump set is producing for a high-value order ( ), and the basic loss rate is according to the table; at the same time, due to the slight blockage of the inlet filter, its operating efficiency decreases, and the measured actual power consumption is , while the theoretical power consumption model shows that the current working condition should be , so ; in addition, the vibration sensor data shows that there is a slight abnormal sound inside, and after sample entropy calculation, the state fluctuation risk multiplier is obtained; the real-time consumption rate at this moment is calculated by the consumption dynamic accounting module as: , wherein, h is hour; this result is 18 times larger than its consumption rate under normal working conditions , so as to accurately quantify the comprehensive loss of business value operation efficiency and stability risk into the health budget consumption of the equipment; then, The budget management module is based on the initial allocation. indivual Total amount and by Real-time consumption calculated by the dynamic accounting module Consumption rate, at a rate not lower than The frequency of continuous updates to the remaining device The remaining credit limit; the decision-driven module then monitors the remaining credit limit in real time; when the remaining credit limit... If the credit limit is lower than a preset warning threshold, for example... When the remaining budget falls below a certain threshold, the decision-driven module is configured to trigger a budget shortfall warning and operations management action, pushing a warning message to the management platform; when the remaining budget further falls below a certain threshold... When the action threshold is reached, a higher-priority operation and maintenance management action is triggered, such as automatically generating an electronic work order requiring inspection or maintenance. Furthermore, to enhance the forward-looking nature of decision-making, the decision-driven module also provides a human-computer interaction interface that responds to user input containing future business scenarios (which determine...). ) and the estimated runtime of the simulation request, and call Consume dynamic accounting modules to predict the remaining device resources under this simulated request. The future consumption trajectory of the quota provides data support for managers to schedule production.

[0034] To address the attribution challenge in operations and maintenance management, namely, only knowing that the risk is high ( In cases where the value surges but the nature of the risk is unknown, this system also includes a volatility pattern analysis module; this module, along with... The computation operates in parallel, utilizing a data sequence of the same high-frequency state data and performing a Fast Fourier Transform (FFT) to extract its frequency domain energy distribution characteristics. The system has a pre-built prototype library storing various typical wave prototypes, such as steady-state switching noise prototypes (characterized by wide bandwidth and transient energy) and typical mechanical imbalance vibration prototypes (characterized by energy concentrated at specific low-frequency harmonics). The wave pattern analysis module performs pattern matching calculations between the real-time extracted frequency domain features and the prototype library, and the classification annotations of the matched typical wave prototypes are output by the decision-driven module as both decision reference information and early warning information. Thus, the information received by managers is no longer isolated risk values, but annotations with business implications, such as... The consumption rate increases instantaneously due to high state fluctuation entropy. The fluctuation mode prototype, namely the steady-state operating condition switching noise, provides a basis for determining whether the event requires immediate intervention. In the technical solution of this invention, a maintenance operation is considered as a supplement to the equipment health budget. To this end, the system also includes a... The supplementary module is configured to: after performing maintenance operations on the fluid measurement and control equipment, determine a corresponding function based on the type of maintenance operation and a predetermined effect verification procedure. Replenishment amount, and used to increase the remaining capacity of the equipment. Credit; for example, for a major overhaul involving the replacement of bearings and seals, supplementary credit can be provided according to regulations. indivual of Credit limit; To enable the system's resource allocation to have adaptive optimization capabilities across cycles, the system further includes an asset behavior profiling analysis module; this module analyzes the assets within a management cycle at the end of that cycle. Consume historical data and Supplement historical data and extract a set of management characteristic indicators that characterize the economic behavior of equipment, such as high... Under working conditions Consumption percentage or maintenance replenishment The utility cycle; subsequently, the asset behavior profiling module uses unsupervised clustering algorithms (such as K-means clustering) to automatically group devices based on these characteristic indicators, determining a behavior profile classification for each device, such as strategic cornerstone type or hidden drainer type; at the beginning of the next management cycle, The quota allocation module will classify devices based on their behavioral profiles and apply a set of differentiated allocation rules, such as increasing the quota for strategic cornerstone devices. initial The total amount will be reduced for devices that cause hidden wear and tear. This allows for the optimization of budget allocation strategies.

[0035] Meanwhile, to enhance the resource allocation flexibility of equipment groups when dealing with localized high-intensity business impacts, the system also introduces a cross-equipment resource integration mechanism; the system includes an asset relationship definition module, used by engineers during the system configuration phase to determine whether there are functional substitutability relationships between equipment based on the process flow diagram. For example, two pumps, A and B, which serve as backups for each other, are defined as a first-level fully substitutable cluster; the system also includes a The budget allocation module is configured to perform internal transfers of funds when an allocation condition is met. The allocation condition is: when the remaining balance of the first device in the cluster (e.g., primary pump A) is... The credit limit is lower than the first preset threshold (e.g.) ), and the remaining second device in the cluster (such as standby pump B) The credit limit is higher than the second preset threshold (e.g.) At this time, the system allows the use of a preset value. to Financing cost coefficient within the range (e.g.) ), part of the second device The credit limit is transferred to the first device; for example, deducted from Pump B's account. indivual Add to Pump A's account indivual This avoids disrupting critical production tasks due to the depletion of a single equipment budget; finally, to enhance the system's ability to cope with the impact of nonlinear strategic opportunities, the system also includes a strategic... The system includes an issuance module and a maintenance liability management module; when a strategic business event occurs in the upper-level management system such as ERP, meeting specific conditions, such as a short-term surge in order volume for a certain product exceeding historical peaks, [the system will take action]. At that time, strategic The issuance module responds to an authorization command issued by a user with the highest administrative privileges (such as the production director) and adds a temporary [application name] to the relevant device. Credit lines were allocated to meet the needs of overcapacity production; meanwhile, the debt management module was configured to maintain... When the credit limit is increased, a future maintenance liability equal to that credit limit is automatically generated in the management system. );Should This represents a corporate commitment to repay the overdraft on equipment lifespan potential during a future business window by performing compensatory, higher-standard maintenance; the maintenance liability management module will, based on this, address planned downtime or off-season production. An automatic compensatory maintenance work order is triggered, thereby ensuring a balance between short-term strategic overdraft and long-term value maintenance.

[0036] Example 1: This example is a specific operational instance of the technical solution in a particular industrial management scenario. All technical features and parameter settings are subject to the disclosure in the specific implementation method. In the production management of a large-scale integrated refining and chemical enterprise, its core hydrocracking unit is scheduled for a week-long annual shutdown overhaul next month. This overhaul is a critical production and operation activity, aiming to ensure that the entire unit can maintain stable production with high overall efficiency and a long operating cycle after the overhaul. Against this backdrop, a core circulating hydrogen compressor unit that supplies high-pressure hydrogen-rich gas to the cracking unit is subject to maintenance loss at the beginning of the year. Based on its design life and this year's production targets, the quota allocation module was allocated 12,000 standard depreciation units. The initial The total amount, one week before this major overhaul, for the compressor unit The budget management module displays its The remaining credit limit is 4500 The situation is stable.

[0037] On Monday morning of that week, a minor leak occurred in an auxiliary pipeline of the compressor unit, causing a slight fluctuation in the unit's outlet pressure. This change in physical condition would not trigger an alarm threshold in a conventional condition-monitoring system; however, in the system of this invention, The dynamic accounting module acquired high-frequency pressure sensor data from the unit and used a sample entropy algorithm to calculate that the complexity of the pressure signal increased compared to the stable operating state, resulting in a higher risk multiplier for state fluctuations. From the benchmark value Rise to Meanwhile, to maintain the constant pressure required for the process flow, the unit's control system automatically increased the operating load, resulting in an increase in the actual power consumption measured by the electricity meter. Slightly increased, while its theoretical power consumption The process efficiency correction factor remained unchanged under this operating condition, thus affecting the efficiency of the process. Depend on Rise to The changes in these two aspects reflect the synergistic effect of the system mechanism; that is, a tiny physical fault source is simultaneously quantified in terms of both stability and efficiency. The multiplicative amplification factor of the consumption rate; at this point, the management data in the Enterprise Resource Planning (ERP) system has marked the coming week as a pre-major overhaul preparation period, and this business process context information is... The dynamic accounting module consumes the data and, according to preset rules, multiplies the business context of the compressor unit. Automatically adjust to a high value The underlying management logic behind this approach is that any damage to the equipment's condition at this moment will directly affect its initial health condition when it enters the overhaul window, thus posing a potential economic risk to its long-term operation after the overhaul; therefore, The real-time consumption of the unit calculated by the dynamic accounting module Consumption rate, in its basic physical loss rate If it remains unchanged, because , and The combined effect of the three multipliers increases, leading to The budget management module's remaining balance began to be depleted at a rate far exceeding normal levels, prompting the system's decision-driven module to issue a warning on Monday afternoon. Warning of rapidly depleting budget.

[0038] The core of this warning message is no longer whether the physical parameters exceed the limits, but rather the equipment itself. The quota is being consumed at an unsustainable high cost; this change shifts the focus of operational decision-making from a device technical issue to an asset budget and risk management issue; instead of facing the trade-off between immediate shutdown to deal with a minor leak and production loss, the manager is now faced with a budget decision of whether to allow the health reserve of the equipment to be continuously consumed at the current high price; in view of this, the manager decides to arrange a short online tightening operation during the production trough period on Tuesday, after the operation is completed, the pipeline leakage is eliminated, and the outlet pressure of the unit is restored to stability; after the operation is completed, The consumption dynamic accounting module updates the calculation results in real time, and the And The values return to The real-time Consumption rate also returns to normal levels; The interface of the budget management module shows that the quota consumption curve becomes flat again, and finally, the compressor unit enters the annual overhaul window with a healthy Budget surplus; by quantitatively coupling physical wear and tear with dynamic business value, operational efficiency, and stability risk, the entire management process can be carried out in a clear economic framework, avoiding potential economic losses due to the lagging or ambiguity of physical indicators.

[0039] Example 2: To objectively verify the effectiveness of the technical solution of the present application in dynamically quantifying the cost of equipment wear and tear under different business value periods, a comparative test is designed; the purpose of the test is to compare the test group using the operation and maintenance management system of the present application with a control group using a traditional state monitoring management method based on physical thresholds under the same, reproducible physical wear and tear process and business scenario simulation, to quantitatively evaluate the differences in decision-making information provided by the two methods; the test platform consists of a 15kW centrifugal fluid pump, a programmable logic controller (PLC), and a corresponding data acquisition system; the platform precisely controls the start and stop, speed, and operating load of the pump to simulate different operating conditions; the data acquisition system includes a three-axis acceleration sensor installed on the pump bearing seat, with a sampling frequency of 2048Hz to obtain high-frequency vibration state data; a digital power meter with an accuracy of 0.5S is used to acquire real-time power consumption data; and a set of upper computer software system is used to record all the data and execute the management logic; it should be noted that in order to simulate a gradual and controllable physical wear and tear process, a small, known mass unbalance body is gradually added to the impeller of the pump to simulate the continuous deterioration of mechanical imbalance.

[0040] The experiment was divided into two independent, sequential phases, each lasting 100 operating hours, corresponding to the control group and the sample group of the present invention, respectively. The management logic for the control group was as follows: the root mean square (RMS) value of the vibration signal was continuously monitored, and when this value exceeded the alarm threshold set according to the ISO10816-3 standard... When a maintenance alarm is triggered, the management logic of the sample group of this invention is as follows: according to the method disclosed in the specific implementation method, the pump is set... Standard loss unit ( The initial Total amount, and continuously calculate in real time. Consumption rate, when remaining The amount is lower than the preset limit. (Right now When the time condition is reached, an early warning is triggered; during the test of the two sample groups, the same operating conditions and fault evolution sequence are applied through the PLC, as follows: From hour 0 to hour 40, normal production conditions are simulated, at which time the business context multiplier... Set as From hour 41 to hour 60, simulate high-value order production conditions. Set as From hour 61 to hour 100, normal production conditions will be restored. Set again The imbalance fault was introduced at the 20th hour and worsened linearly over time. During the test of the control group, the system continuously recorded vibration data until the 85th hour, when the pump's vibration RMS value reached [value missing]. When the physical alarm threshold is exceeded for the first time, the system triggers a maintenance alarm. Prior to this, although the physical wear and tear of the equipment had accelerated due to the continued development of the fault during the high-value production period from the 41st to the 60th hour, the traditional management method failed to provide any early warning information because the physical quantity had not yet reached the absolute threshold, resulting in decision-makers being unable to perceive the loss process with higher economic risks that occurred during this critical business window.

[0041] During the testing of the sample group of this invention, the system, from hour 0 to hour 40, The consumption rate remained at a low baseline level; when the test entered the 41st hour, although the vibration RMS value at this point was only... It is far below the physical alarm threshold, but because Value by leap to Meanwhile, the escalating faults have led to a multiplier of risk for state fluctuations. Process efficiency correction factor Rise to and , The real-time The consumption rate instantaneously increased by more than 20 times; this change led to The quota margin in the budget management module began to rapidly decrease, and at the 52nd hour, the remaining The quota was first lower than The system decision driving module correspondingly triggered the budget shortage warning; Table 1 is a comparison of the state data of the two groups at several key time points.

[0042] Table 1: Comparison of test data.

[0043] Key time points (hours) Vibration RMS value (mm / s) System state Control system action Inventive system action 40 2.7 Before entering high value operating condition No No 52 3.1 During high value operating condition No Trigger ODR budget alert 60 3.3 Exiting high value operating condition No Continue alerting 85 4.52 During normal operating condition Trigger physical threshold alert Continue alerting

[0044] The test data show that, referring to Table 1, the management system of the inventive group can provide early warning information about the deterioration of the economic health state of the equipment to the manager 33 hours earlier (the 85th hour compared to the 52nd hour) by introducing a mechanism that dynamically accounts for physical wear and tear, business value, operating efficiency, and stability risk, compared to the traditional management method of the control group; the triggering time of the warning corresponds to the time window in which the economic cost of equipment wear and tear is amplified due to changes in business context, thereby providing a decision basis for preventive intervention and avoiding bearing asset value loss during high-value production, and the technical effect has been objectively verified.

[0045] To further verify the decision support advantage of the inventive system compared to the traditional technology that only integrates periodic maintenance and state monitoring when responding to sudden high-load business impact, the following Comparative Example 1 is additionally designed.

[0046] Comparative Example 1: This comparative example aims to simulate a common industrial scenario: a device that is at the end of the maintenance period, has potential wear and tear although the physical state has not reached the absolute alarm threshold, and when facing a sudden and exceeding the regular design load production task, the difference in decision support capability provided by different operation and maintenance management systems; the test platform and parameter criteria used in this comparative example are completely consistent with Example 2 except for the management logic, i.e., a centrifugal fluid pump with a rated power of 15 kW is used, and the physical alarm threshold of the vibration root mean square (RMS) value is set to 4.5 mm / s according to the ISO10816-3 standard; the control management method is a common conventional technical path in the field that combines fixed-period maintenance and physical threshold alarm: the maintenance period of the pump is set to 2000 operating hours, and when the operating time reaches or the vibration RMS value exceeds the limit, a maintenance work order is triggered; at the beginning of this comparative example, the pump has accumulated 1800 hours of operation (i.e., at the 90% node of its maintenance period), and the bearing vibration RMS value is 3.8 mm / s, which does not reach the alarm threshold, and the system state is normal.

[0047] The test simulates an emergency order with high profit but must be completed within 48 hours, which requires the pump to run at 120% of its rated load continuously, the management logic of the sample group of the application is the same as that of example 2, the initial ODR total quota is 1000 standard deduction units (SDU), and the warning threshold is 300 SDU remaining, at the beginning of the test, the ODR quota of the sample group of the application is 450 SDU, because the emergency order is of high value, the business context multiplier BCM of the sample group of the application is set to 15; the test process and result data are shown in table 2, under the management of the conventional technology path, because the cumulative running time of the pump does not reach 2000 hours, and the vibration value does not exceed the static threshold of 4.5 mm / s for most of the task execution time, the system does not provide any warning, until the 48th hour, the bearing of the pump fails catastrophically due to overload accelerated fatigue, and the vibration value soars, at this time, the system triggers the physical threshold alarm, but it cannot avoid unplanned downtime and production loss; in contrast, at the beginning of the emergency task, the ODR consumption dynamic accounting module of the sample group of the application integrates high business value (BCM=15), efficiency decline caused by overload (PEM rises to 1.3) and increase of running instability (CVE rises to 2.0) and other factors, calculates the real-time ODR consumption rate, and predicts that the ODR quota will be exhausted within 10 hours, and immediately issues a high-level asset health budget warning that the ODR quota will be exhausted soon, which provides a 47-hour decision window for the manager to make a quantitative decision between executing high-value orders and bearing foreseeable equipment failure risks, and takes risk avoidance measures such as enabling backup equipment or negotiating with customers for batch delivery.

[0048] Table 2: Comparison of test data of comparative example 1.

[0049]

[0050] The test results of the comparative example objectively show that the conventional technology path lacks the core link of quantitatively correlating the physical wear of the equipment with the dynamically changing business value, running efficiency and stability risk in its management mechanism, and has inherent defects of lagging decision information and failing to reveal potential economic risks when dealing with such sudden and high-risk production scenarios.

[0051] Example 3: This example combines Figures 1 to 3 to explain a fluid measurement and control equipment full life cycle operation and maintenance management system based on multi-source data, such as Figure 1As shown, the system uses four external data sources as inputs: an Enterprise Resource Planning (ERP) / Manufacturing Execution System (MES) for management data, a digital energy meter for real-time power consumption data, an equipment operation data source for basic physical loss data, and a vibration / acoustic sensor for high-frequency status data. The quota allocation module allocates initial quotas to devices based on management data. Total amount, The dynamic consumption calculation module integrates information from four types of data sources to calculate real-time consumption. Consumption rate The budget management module is based on the initial Total amount, real-time Consumption rate and by The supplementary module is provided after maintenance. Supplement amount, for the remaining The credit limit is continuously updated; the decision-driven module updates the credit limit based on the remaining amount. The system analyzes credit limit status and, combined with decision references (categorized annotations) generated from high-frequency status data by the fluctuation pattern analysis module, ultimately outputs operational management actions such as alerts and work orders. Additionally, the system includes a human-computer interface for receiving user simulation requests and invoking the decision-driven module for future consumption prediction, as well as an asset behavior profiling module. The budget management module provides historical data. The data generates differentiated allocation rules and feeds them back to... The quota allocation module thus forms a complete closed-loop management and optimization process.

[0052] like Figure 2 As shown, the operation and maintenance management terminal includes a human-machine interface web front-end. Users initiate access and simulation requests to the operation and maintenance management application server through this terminal. This server contains a core processing engine and analysis service module, responsible for executing the system's core calculations and logic. On the one hand, the operation and maintenance management application server communicates with the enterprise application server, which carries the enterprise resource planning / manufacturing execution system, through a business management data interaction interface. On the other hand, it reads and writes data to the data storage system, which includes real-time / time-series databases and relational databases. At the device end, signals collected by sensors / instruments are sent to the data acquisition unit / edge gateway. The data acquisition agent in this unit obtains data from the fluid measurement and control equipment through the industrial bus and sends the processed real-time device data to the operation and maintenance management application server. This constitutes a complete physical deployment from data acquisition, processing and analysis to user interaction.

[0053] like Figure 3 As shown, the asset relationship definition module first sends... The budget coordination module provides equipment substitution relationships. The budget financing module then proceeded to The budget management module queries the remaining credit limits for device A and device B. After receiving feedback that device A has insufficient credit and device B has sufficient credit, if the conditions for credit transfer are met, The budget allocation module applies an allocation cost coefficient to deduct the credit limit from device B and increase the credit limit for device A, while simultaneously calling... The budget management module updates the balances of device B and device A respectively, and after the financing is completed, the decision-driven module sends the financing transaction record to the administrator; if the financing conditions are not met, the decision-driven module suggests other operation and maintenance strategies to the administrator.

[0054] Example 4: This example aims to standardize the offline calibration and construction procedures for the key parameters and internal model of the system, so as to eliminate the uncertainty in specific engineering application deployment and ensure that the decision logic of the entire system has a reproducible and auditable basis. In a specific engineering deployment scenario, when the system of the present invention needs to be applied to a newly installed key raw material delivery pump for a continuous polymerization reactor, the initial state of the pump is that the mechanical installation and electrical connection have been completed, and it has a multi-channel historical dataset with precise timestamps containing at least 1000 hours of its initial trial operation. The data source of this dataset is an integrated data acquisition unit, whose functional specifications are: able to synchronously record the pump's inlet pressure, outlet pressure, instantaneous flow rate, motor input power, and triaxial vibration signal at the bearing housing, wherein the acquisition accuracy of power data is not less than 0.5%, and the sampling frequency of vibration data is not less than 2048Hz. In addition, the dataset is also associated with the business process context label corresponding to each moment through the interface with the Manufacturing Execution System (MES), which can be regular production, product brand switching, or standby.

[0055] Given this initial state, establish the theoretical power consumption of the pump. The baseline model follows these procedures: First, select all data segments from the historical dataset that are in a stable operating state, excluding processes involving start-up, shutdown, or drastic changes in operating conditions; second, calculate the hydraulic power-to-electric power conversion efficiency for each set of records in the selected data segments, and select the one with the highest conversion efficiency. The data points are used as the optimal operating condition dataset for the pump; finally, using this optimal operating condition dataset, a multivariate nonlinear regression analysis method is employed to establish a model with real-time flow rate and outlet pressure as inputs and theoretical power consumption as the input. The output mathematical model is embedded in the system and used as an efficiency correction factor in subsequent online computation processes. The baseline; next, the business context multiplier. Calibration is performed, and the calibration procedure is as follows: First, in collaboration with the production management and finance departments, a quantitative assessment is conducted on the economic losses caused by unplanned downtime per unit time for the pump under different business process contexts, denoted as... The calculation yielded the following results regarding the normal production context: The loss is defined as 10,000 yuan / hour; however, during the product brand switching window, downtime will result in the scrapping of an entire batch of raw materials and delays in production plans, and the calculated loss is significant. The first step is to calculate the cost per hour as 10,000 yuan; the second step is to analyze the context of each business process. The value is determined as the ratio of the economic loss in this context to the baseline loss, i.e.: According to this formula, the pump in normal production Value And when switching product brands The value is then labeled as This procedure transforms a management concept into a traceable engineering parameter supported by specific calculation formulas and data sources.

[0056] Finally, to enable the system's fluctuation pattern analysis module to attribute and annotate risks, a fluctuation pattern prototype library needs to be established. Using the aforementioned historical dataset, vibration data segments corresponding to events such as routine production, product brand switching, and known minor blockage of the pump inlet filter are extracted. For each data segment, its frequency domain energy distribution characteristics are extracted using Fast Fourier Transform (FFT), and these feature vectors, along with the corresponding event labels, are stored in the prototype library. The steady-state vibration spectrum characteristics under routine production conditions are marked as the steady-state operation prototype. Through this series of procedures, the core model, key parameters, and knowledge base of the operation and maintenance management system, originally general-purpose software, have all undergone personalized calibration for data-driven operation of this specific equipment. The system now possesses all the initial conditions for online operation, and subsequent... Both calculations and decision-making outputs will be based on this set of objectively calibrated, logically transparent internal criteria.

[0057] Example 5: This example aims to standardize the offline calibration and construction procedures for the key parameters and internal model of the system, so as to eliminate the uncertainty in specific engineering application deployment and ensure that the decision logic of the entire system has a reproducible and auditable basis. In a specific engineering deployment scenario, when the system of the present invention needs to be applied to a newly installed key raw material delivery pump for a continuous polymerization reactor, the initial state of the pump is that the mechanical installation and electrical connection have been completed, and it has a multi-channel historical dataset with precise timestamps containing at least 1000 hours of its initial trial operation. The data source of this dataset is an integrated data acquisition unit, whose functional specifications are: able to synchronously record the pump's inlet pressure, outlet pressure, instantaneous flow rate, motor input power, and triaxial vibration signal at the bearing housing, wherein the acquisition accuracy of power data is not less than 0.5%, and the sampling frequency of vibration data is not less than 2048Hz. In addition, the dataset is also associated with the business process context label corresponding to each moment through the interface with the Manufacturing Execution System (MES), which can be regular production, product brand switching, or standby.

[0058] Given this initial state, establish the theoretical power consumption of the pump. The baseline model follows these procedures: First, select all data segments from the historical dataset that are in a stable operating state, excluding processes involving start-up, shutdown, or drastic changes in operating conditions; second, calculate the hydraulic power-to-electric power conversion efficiency for each set of records in the selected data segments, and select the one with the highest conversion efficiency. The data points are used as the optimal operating condition dataset for the pump; finally, using this optimal operating condition dataset, a multivariate nonlinear regression analysis method is employed to establish a model with real-time flow rate and outlet pressure as inputs and theoretical power consumption as the input. The output mathematical model is embedded in the system and used as an efficiency correction factor in subsequent online computation processes. The baseline; next, the business context multiplier. The calibration procedure is as follows: First, in collaboration with the production management and finance departments, a quantitative assessment of the economic losses caused by unplanned downtime of the pump per unit time under different business process contexts is performed, denoted as... The calculation yielded the following results regarding the normal production context: The loss is defined as 10,000 yuan / hour; however, during the product brand switching window, downtime will result in the scrapping of an entire batch of raw materials and delays in production plans, and the calculated loss is significant. The first step is to calculate the cost per hour as 10,000 yuan; the second step is to analyze the context of each business process. The value is determined as the ratio of the economic loss in this context to the baseline loss, i.e.: ,in, Representative at the Business context multiplier under a business process context; : Represents the first Economic losses caused by unplanned downtime per unit time in the context of various business processes; This represents the economic loss in the specific context of routine production, and this value is defined as the baseline loss, according to this formula, for the pump during routine production. Value And when switching product brands The value is then labeled as Finally, to enable the system's fluctuation pattern analysis module to attribute and annotate risks, a fluctuation pattern prototype library needs to be established. Using the aforementioned historical dataset, vibration data segments corresponding to events such as routine production, product brand switching, and known minor blockage of the pump inlet filter are extracted. For each data segment, its frequency domain energy distribution characteristics are extracted using Fast Fourier Transform (FFT), and these feature vectors, along with the corresponding event labels, are stored in the prototype library. The steady-state vibration spectrum characteristics under routine production conditions are marked as steady-state operation prototypes. Through this series of procedures, the core model, key parameters, and knowledge base within the system have all been data-driven and personalized calibrated for this specific equipment. The system now possesses all the initial conditions for online operation.

[0059] Example 6: This example aims to provide a standardized annual calibration and optimization procedure for ensuring the timeliness and adaptive reconstruction of the system's internal models and decision thresholds during long-term operation, in order to cope with the natural evolution of equipment performance and dynamic changes in the business environment. In an asset management scenario, after the system of this invention has been running stably on a set of key centrifugal compressors for a complete annual cycle, in order to ensure its management effectiveness in the following year, the system automatically triggers an annual verification and reconstruction process. The initial input of this process is all the timestamped operating data of the compressor set recorded in the past year, the associated business context tags, and all maintenance depreciation rights. Consumption and replenishment of historical flow data; the first step in the process is to analyze the theoretical power consumption. The effectiveness of the benchmark model is verified by automatically calculating the results under all stable operating conditions over the past year. The theoretical power consumption predicted by the model and the actual power consumption recorded by the electricity meter The root mean square error (RMSE) between the values ​​exceeds a preset threshold. If the threshold is reached, it indicates that the compressor's energy efficiency characteristics have drifted, and the system will automatically call [a new feature / method]. The model building procedure utilizes the best operating condition dataset from the most recent year. The model is then refitted and updated.

[0060] Secondly, the process multiplier of business context Upon review of the matrix, the system sends a review request to the designated production and finance administrators, which lists all current business context tags and their corresponding economic losses per unit time for unplanned downtime. The administrator, based on the new year's production plan and product pricing system, adjusts the values ​​for each item. The value will be reviewed and updated; after the update is complete, the system will... The calculation formula will automatically use the new one. Value and updated baseline loss Recalculate and overwrite the original Matrix; furthermore, for the decision-driven module The budget warning threshold has been transformed from a static setpoint into a dynamic parameter driven by historical data. The system executes an adaptive threshold optimization procedure that analyzes all events that caused unplanned downtime in the past year and extracts the remaining data from the 72 hours prior to these events. By statistically analyzing the credit limit change trajectory, a mechanism can be calculated that provides at least 48 hours of early warning time in 95% of cases. A percentage of the credit limit, and this percentage, for example The threshold for early warning in the following year is set; the annual process also extracts and solidifies management characteristic indicators used for profile clustering; the system extracts data from each compressor unit. From historical transaction data, a set of specific quantitative indicators are calculated and extracted, including: indicators representing volatility consumption. The Gini coefficient, representing the intensity of undertaking critical tasks, indicates the high level of daily consumption. Operating conditions Consumption percentage, and the unit representing maintenance responsiveness. The supplementary data supports the average number of fault-free operating days; these metrics are used as input to an unsupervised clustering algorithm to generate a new year's asset behavior profile classification that reflects the economic behavior of the equipment in the previous cycle; by executing this complete annual calibration and optimization procedure, the system updates its core data model, management parameter matrix, decision logic thresholds, and analysis feature set without interrupting online monitoring tasks, in order to adapt to the long-term dynamic changes of equipment and business environment.

[0061] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0062] Finally, it should be noted that the above examples are merely intended to illustrate the technical solutions of the present application and not to limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the present application.

Claims

1. A multi-source data based fluid measurement and control device full life cycle operation and maintenance management system, characterized in that, The system comprises: an ODR quota allocation module configured to allocate an initial total ODR quota to a fluid measurement and control device at the beginning of a management cycle; an ODR consumption dynamic accounting module configured to obtain operation data reflecting physical loss of the device to determine a basic physical loss rate, obtain management data reflecting the context of the business process of the device to determine a business context multiplier, obtain real-time actual power consumption and theoretical power consumption of the device to determine a process efficiency correction factor, and obtain high-frequency state data of the device and calculate the information entropy of the high-frequency state data within a time window to determine a state fluctuation risk multiplier; the ODR consumption dynamic accounting module is further configured to calculate the real-time ODR consumption rate of the device based on the product of the basic physical loss rate, the business context multiplier, the process efficiency correction factor, and the state fluctuation risk multiplier; the ODR consumption dynamic accounting module is further configured to determine the value of the business context multiplier according to a preset rule that maps management data to the business context multiplier; the management data includes production plan information, order value information, and criticality level information of the fluid measurement and control device in the process flow; an ODR budget management module configured to continuously update the remaining ODR quota of the device according to the initial total ODR quota and the real-time ODR consumption rate; a decision driving module configured to trigger one or more predefined operation and maintenance actions when the remaining ODR quota meets one or more pre-set decision conditions. 2.The multi-source data based fluid measurement and control device whole life cycle operation and maintenance management system according to claim 1, wherein, The system further comprises an ODR supplement module configured to determine an ODR supplement corresponding to a maintenance operation according to the type of the maintenance operation and a predetermined effect verification procedure, and to increase the remaining ODR quota of the device after the maintenance operation is performed on the fluid measurement and control device. 3.The multi-source data based fluid measurement and control device whole life cycle operation and maintenance management system of claim 1, wherein, The ODR consumption dynamic accounting module is further configured to: acquire a real-time electric energy meter reading of the device as an actual power consumption; and based on design performance curve data or historical optimal working condition data of the device, establish a benchmark model mapping a current working condition to a theoretical power consumption, and determine the theoretical power consumption by the benchmark model; wherein the process efficiency correction factor is determined in the following manner wherein, is the actual power consumption, is the theoretical power consumption.

4. The fluid measurement and control device full life cycle operation and maintenance management system based on multi-source data according to claim 1, characterized in that, The determination of the state fluctuation risk multiplier in the ODR consumption dynamic accounting module is achieved by the following method: obtaining a data sequence of high-frequency state data from a vibration sensor or an acoustic sensor of the fluid measurement and control device; using a sample entropy algorithm to calculate the complexity of the data sequence within a time window, and normalizing the calculation result to generate a scalar value quantifying the uncertainty of the running state of the device as the state fluctuation risk multiplier; wherein the length of the time window and the calculation step are pre-set.

5. The fluid measurement and control device full life cycle operation and maintenance management system based on multi-source data according to claim 4, characterized in that, The system further comprises a fluctuation pattern analysis module configured to perform fast Fourier transform on the data sequence of high-frequency state data in parallel to extract its frequency energy distribution characteristics; performing pattern matching calculation on the frequency energy distribution characteristics and a prototype library storing multiple typical fluctuation prototypes; the decision driving module is further configured to output the classification annotation of the matched typical fluctuation prototype as decision reference information when triggering the operation and maintenance action. 6.The multi-source data based fluid measurement and control device whole life cycle operation and maintenance management system of claim 1, wherein, The decision driving module is further configured to provide a human-machine interaction interface, which is responsive to a simulation request inputted by a user, and the simulation request comprises a future business scenario and a running time length, and the decision driving module is further configured to invoke the ODR consumption dynamic accounting module to predict a future consumption trajectory of the remaining ODR quota of the equipment under the simulation request.

7. The fluid measurement and control device full life cycle operation and maintenance management system based on multi-source data according to claim 1, characterized in that, The system further comprises an asset behavior portrait analysis module, which is configured to extract a set of management feature indexes representing economic behaviors of the equipment based on ODR consumption history data and ODR replenishment history data of the fluid measurement and control equipment in one or more historical management cycles, and to automatically group the equipment according to the management feature indexes by using an unsupervised clustering algorithm, and to determine a behavior portrait classification for each equipment; wherein the ODR quota allocation module applies a set of differentiated allocation rules according to the behavior portrait classification of the equipment. 8.The multi-source data based fluid measurement and control device whole life cycle operation and maintenance management system of claim 1, wherein, The system further comprises an asset relationship definition module configured to determine that there is a functional substitutability relationship between the first fluid measurement and control equipment and the second fluid measurement and control equipment, and an ODR budget refinancing module configured to transfer part of the ODR quota of the second fluid measurement and control equipment to the ODR quota of the first fluid measurement and control equipment according to a preset refinancing cost coefficient in the range of 1.05 to 1.3 when the refinancing conditions are met, i.e., the remaining ODR quota of the first fluid measurement and control equipment is lower than a first preset threshold, and the remaining ODR quota of the second fluid measurement and control equipment is higher than a second preset threshold. 9.The multi-source data based fluid measurement and control device whole life cycle operation and maintenance management system of claim 1, wherein, The system further comprises a strategic ODR issuance module configured to increase a temporary ODR credit quota for the fluid measurement and control equipment in response to an authorization instruction when a strategic business event meeting the conditions is identified, and a maintenance liability management module configured to generate a future maintenance liability associated with the ODR credit quota when the ODR credit quota is increased, and to trigger a compensatory maintenance work order based on the future maintenance liability in a future business window period.

Images