KR102315193B1 - Inverter inspection system for pump - Google Patents

Abstract

The present invention relates to an inverter inspection system for a pump, which includes: a board stage inspection unit which inspects whether each board operates normally at the board level during the manufacture of a pump inverter; and a finished product inspection unit which detects a product error in a finished product assembly stage of the pump inverter after the inspection by the board stage inspection unit.

Description

Inverter inspection system for pump

The present invention relates to an inverter inspection system for a pump.

In general, a pump is a machine that receives mechanical energy from a driving motor, transfers this energy to a liquid to be handled, and discharges the liquid from a low-pressure part to a high-pressure part. In addition, in general, a booster pump is a pump installed in a water supply facility or a high-rise building to compensate for the shortage for obtaining a required water pressure, and may be used alone, but mainly a booster pump consisting of a plurality of booster pumps connected in a parallel structure. The system is being used. When supplying water to a target place using the booster pump system, maintaining a constant pressure of the water supply is very important in the booster pump system. In order to maintain a constant supply pressure, a pump rotation speed control method using an inverter or a pump number control method is used.

On the other hand, the inverter refers to a device that converts DC power into AC power of a size and frequency desired by a user. As a method of inspecting this performance, an efficiency test comparing inverter input power and output power is mainly used.

However, when manufacturing an inverter for a conventional pump, there is a problem that it takes a lot of time for functional inspection and testing by applying a system that detects product errors only after assembling the finished product, even though there is a minor defect at the bore level and can be easily corrected. In addition, in the prior art, for the inverter performance test, the inspector directly inputs the inverter set value and the instrument set value, drives the inverter, and then checks the measured data to determine whether the inverter is operating normally. When inputting a set value, there was a problem that an accurate performance check could not be performed.

The present invention has been devised to solve the above problems,

From the board stage before assembling the finished product of the pump inverter, the function is inspected through the test jig unit to block even minor defects at the source to reduce the overall inspection time, and the construction of the inspection standard generation unit and judgment unit applied with artificial intelligence enables accurate and quick finished products We provide an inverter inspection system for pumps implemented so that performance inspection can be performed automatically.

The technical problems of the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the following description.

A pump inverter inspection system according to an embodiment of the present invention includes: a board stage inspection unit for inspecting whether each board operates normally at the board level when the pump inverter is manufactured; and a finished product inspection unit that detects a product error in the finished product assembly stage of the pump inverter after the inspection by the board stage inspection unit.

In one embodiment, the board stage inspection unit, the corresponding switch for signal detection is turned on according to the switch control signal, and a test jig unit for detecting the inspection signal from the measurement target board through the turned-on switch; a frequency counter unit for counting the frequency of the detection signal output from the test jig unit; a signal input/output unit for inputting and outputting signals for controlling switches for signal detection of the test jig unit; a signal processing unit converting the test signal detected by the test jig unit into digital data; a main control unit that processes data input from the signal processing unit and the frequency counter unit by inspection/adjustment software, and outputs a switch control signal of the test jig unit according to the result to the signal input/output unit; and a display unit for outputting an adjustment item selection signal and an adjustment reference value to the main control unit, and receiving and displaying display data from the main control unit.

In an embodiment, the signal processing unit may include: a first selection signal output unit for selecting and outputting one of two signals detected by the test jig unit and inputted from a first channel and a second channel; a first data conversion unit converting the signal output from the first selection signal output unit into data; a sub-control unit which processes the signal data output from the data conversion unit by inspection/adjustment software and outputs text data according to the result; a signal processing unit receiving the synchronization signal output from the data conversion unit and outputting an enable signal; an address counter receiving the enable signal output from the signal processing unit and outputting address data; a memory unit receiving and storing character data from the sub-control unit, receiving address data output from the address counter, and outputting data stored in the address; a second data conversion unit for outputting the data output from the memory unit as an analog signal; and a second selection signal output unit that selects one of the first selection signal output unit and the second data conversion unit and outputs an input signal to the display.

In one embodiment, the finished product inspection unit, inspection information storage unit for storing the inspection step and criteria for the inverter performance inspection; an inspection standard generation unit for receiving an inspection step and an inspection criterion matching the inspection execution signal from the inspection information storage unit, and analyzing the received information to generate a new criterion; and a determination unit that receives the state data of the inverter and determines whether the inverter is in a normal state or a bad state by comparing it with a predicted result value predicted in response to the new inspection criterion generated by the inspection criterion generating unit. can do.

In one embodiment, the inspection standard generation unit and the determination unit, generate a data set that allows the pump to be optimally maintained at a constant pressure by linking the information about the inverter inspection for the pump and the pressure information of the pump stored in the inspection information storage unit It can be characterized as

In one embodiment, the learning of the artificial neural network is performed by applying the data set to a deep learning algorithm structure including a convolutional neural network and a fully connected deep neural network, and gradually changing the weight of the neural network structure corresponding to the error. It may be characterized in that the result is fed back and learned through a backpropagation algorithm.

In one embodiment, the inspection criterion generating unit, the set pressure of the pump system to which the inverter for the pump is connected, the set number of pump rotations, the discharged flow rate, the discharged pressure, the maximum flow rate, the maximum efficiency flow rate, the input voltage and current of the inverter, the inverter output voltage can be used as input data, and it can be detected by inputted inverter output voltage and voltage measuring device. Calculating the output power using the DC voltage information, estimating the discharge flow rate of each pump using the set rotational speed and output power and the discharged pressure connected to each of the pump inverters; Calculating the number of rotations of the pump according to the set pressure and the discharged pressure, the step of the pump inverter estimating the maximum flow rate and the maximum efficiency flow rate based on the set pressure of the pump system and the performance curve data of the pump system, the pump inverter receives the estimated discharge flow rate of each pump and calculates the total flow. Calculating and estimating the required rotational speed of the pump to be driven at the optimal efficiency operating point, and determining whether the inverter for the pump can consistently satisfy the calculated and estimated required rotational speed of the pump for a long time The performance of the pump system It characterized in that it comprises the step of determining whether the output power does not exceed a maximum of 22 kW by weighting the curve data and the output power of the inverter for the pump, and then the state of the inverter for the pump to be inspected by the determination unit 230 is normal If it is determined whether it falls within the range, and the output power exceeds the maximum of 22kW or is determined to be in a defective state, the inspection standard generator analyzes the defective values and changes the voltage value input to the pump inverter through feedback in 1V units, It may be characterized in that a new test data standard is generated after each of the above steps is repeated.

The above-described problem solving means are merely exemplary, and should not be construed as limiting the present application. In addition to the exemplary embodiments described above, additional embodiments may exist in the drawings and detailed description.

According to one aspect of the present invention described above, the overall inspection time can be reduced by detecting the source of defects through the board stage inspection unit configuration, and the pump can accurately and optimally control the pressure of the pump through the finished product inspection unit to which artificial intelligence is applied. It can provide the effect that the inverter for the finished product can be manufactured.

The effects of the present invention are not limited to the above-mentioned effects, and various effects may be included within the range apparent to those skilled in the art from the description below.

1 is a block diagram showing the configuration of an inverter inspection system for a pump according to an embodiment of the present invention.
2 is a view showing a board stage inspection unit of the inverter inspection system for a pump according to an embodiment of the present invention.
3 is a diagram illustrating a signal processing unit of an inverter inspection system for a pump according to an embodiment of the present invention.
4 is a view showing a finished product inspection unit of the inverter inspection system for a pump according to an embodiment of the present invention.
5 is a view showing an inspection procedure of the finished product inspection unit according to an embodiment of the present invention.
6 is a graph showing the performance curve data of the pump.
7 is a diagram illustrating a notification unit according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0023] Reference is made to the accompanying drawings, which show by way of illustration specific embodiments in which the present invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the present invention. It should be understood that the various embodiments of the present invention are different but need not be mutually exclusive. For example, certain shapes, structures, and characteristics described herein with respect to one embodiment may be implemented in other embodiments without departing from the spirit and scope of the invention.

In addition, it should be understood that the location or arrangement of individual components within each disclosed embodiment may be changed without departing from the spirit and scope of the present invention. Accordingly, the detailed description set forth below is not intended to be taken in a limiting sense, and the scope of the invention, if properly described, is limited only by the appended claims, along with all scope equivalents to those claimed. Like reference numerals in the drawings refer to the same or similar functions throughout the various aspects.

On the other hand, throughout this specification, when a part "includes" a certain component, this means that the specifically opposite description ?A does not exclude one other component, but may further include other components. do. In addition, as used herein, a “unit” for a component performs at least one function or operation. And “unit” may perform a function or operation by hardware, software, or a combination of hardware and software.

Terms including an ordinal number, such as first, second, etc., may be used to describe various elements, but these elements are not limited by the above-described terms. The above terminology is used only for the purpose of distinguishing one component from another component.

In the present specification, the term "comprising" is intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, and includes one or more other features, number, or step. , it should be understood that it does not preclude in advance the possibility of the existence or addition of an operation, component, part, or combination thereof. When a component is referred to as being “connected to” another component, it may be directly connected or connected to the other component, but it should be understood that another component may exist in between.

In addition, in describing the present invention, when it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the gist of the present invention, the detailed description thereof will be abbreviated or omitted.

Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the drawings.

1 is a block diagram showing the configuration of an inverter inspection system for a pump according to an embodiment of the present invention.

2 is a view showing a board stage inspection unit of the inverter inspection system for a pump according to an embodiment of the present invention.

3 is a diagram illustrating a signal processing unit of an inverter inspection system for a pump according to an embodiment of the present invention.

4 is a view showing a finished product inspection unit of the inverter inspection system for a pump according to an embodiment of the present invention.

5 is a view showing an inspection procedure of the finished product inspection unit according to an embodiment of the present invention.

6 is a graph showing the performance curve data of the pump.

7 is a diagram illustrating a notification unit according to an embodiment of the present invention.

Referring to Figure 1,

The inverter inspection system 10 for a pump according to an embodiment of the present invention may include a board stage inspection unit 100 and a finished product inspection unit 200 .

The board stage inspection unit 100 may inspect whether each board operates normally at a board stage such as a control board and a power board when the inverter for the pump is manufactured.

The finished product inspection unit 200, after the inspection by the board stage inspection unit 100 is passed, when the finished product of the pump inverter is assembled, by performing a complete inspection on all products, it is possible to prevent the occurrence of defects and realize quality maintenance.

Referring to Figure 2,

In one embodiment, the board step inspection unit 100 according to an embodiment includes a test jig unit 110 , a frequency counter unit 120 , a signal input/output unit 130 , a signal processing unit 140 , and a main control unit 150 ). and a display unit 160 .

The test jig unit 110 may turn on a corresponding switch for signal detection according to the switch control signal, and detect a test signal from the measurement target board through the turned on switch. The test jig may include various test jigs such as a control board test jig and a power board test jig.

The frequency counter unit 120 may count the frequency of the detection signal output from the test jig unit 110 .

The signal input/output unit 130 may input/output signals for controlling switches for signal detection of the test jig unit 110 .

The signal processing unit 140 may convert the test signal detected by the test jig unit 110 into digital data.

The main control unit 150 processes the data input from the signal processing unit 140 and the frequency counter unit 120 by the inspection/adjustment software, and transmits the switch control signal of the test jig unit 110 according to the result to the signal input/output unit. (130) can be output.

The display unit 160 may include a display unit that outputs an adjustment item selection signal and an adjustment reference value to the main control unit 150 , and receives and displays display data from the main control unit 150 .

In more detail, the main control unit 150 receiving the data for the selection item may output the inspection/adjustment item selection control signal to the signal input/output unit 130 . The inspection/adjustment item selection control signal output from the main control unit 150 is input to the test jig unit 110 through the signal input/output unit 130, the corresponding switch is turned on, and the signal detected by the jig connected to the switch is displayed. The signal may be output to the signal processing unit 140 or the frequency counter unit 120 . On the other hand, the switch, for example, a relay, a thyristor, such as switches that are turned on/off by an electric control signal may be used.

Referring to Figure 3,

The signal processing unit 140 according to an embodiment includes a first selection signal output unit 141 , a first data conversion unit 143 , a sub-control unit 145 , a signal processing processing unit 146 , an address counter 147 , It may include a memory unit 148 , a second data conversion unit 144 , and a second selection signal output unit 142 .

The first selection signal output unit 141 may select and output one of the two signals detected by the test jig unit 110 and inputted from the first channel and the second channel.

The first data conversion unit 143 may convert the signal output from the first selection signal output unit 141 into data.

The sub-control unit 145 may process the signal data output from the first selection signal output unit 141 by the inspection/adjustment software and output character data according to the result.

The signal processing unit 146 may receive the synchronization signal output from the first data conversion unit 143 and output an enable signal.

The address counter 147 may receive the enable signal output from the signal processing unit 146 and output address data.

The memory unit 148 may receive and store character data from the sub-control unit 145 , and may receive address data output from the address counter 147 and output data stored at the address.

The second data conversion unit 144 may output data output from the memory unit 148 as an analog signal.

The second selection signal output unit 142 may select one of the first selection signal output unit 141 and the second data conversion unit 144 and output an input signal to the display.

More specifically, the detection signal input to the signal processing unit 140 is input to the first selection signal output unit 141 through the first channel or the second channel, and then is selected and output. The signal output from the first selection signal output unit 141 is digitally converted by the first data conversion unit 143 and is input to the sub-control unit 145 . The sub-control unit 145 processes the signal data output from the first data conversion unit 143 by the inspection/adjustment software, and outputs data according to the result to the memory unit 148 .

In addition, the first data conversion unit 143 extracts the synchronization signal from the signal output from the first selection signal output unit 141 and outputs it to the signal processing unit 146 . The signal processing unit 146 receives the synchronization signal and outputs an enable signal to the address counter 147 , and the address counter 147 sequentially generates a memory address from address 0 and outputs it to the memory unit 148 . The memory unit 148 receiving the memory address outputs the data stored in the address to the second data conversion unit 144 . That is, the data processed and stored by the inspection/adjustment software is output to the second data conversion unit 144 . The second selection signal output unit 142 displays the signal detected by the test jig of the test jig unit 110 output from the first selection signal output unit 141 according to the control signal output from the main control unit 150 . Output or output as data.

4 and 5,

In an embodiment, it may include a finished product inspection unit 200 , an inspection information storage unit 210 , an inspection standard generation unit 220 , and a determination unit 230 according to an embodiment.

The inspection information storage unit 210 stores the inspection steps and standards for the inspection of the finished product of the inverter for the pump. It may be stored in the examination information storage unit 210 in a table structure to provide various examination environments. For example, various values such as acceleration time, deceleration time, input current and voltage information, output current and voltage information, horsepower information, capacity information, frequency information, and output waveform information of the inverter may be stored.

The inspection standard generating unit 220 may receive an inspection step and an inspection standard matching the inspection execution signal from the inspection information storage unit 210 , and may generate a new criterion by analyzing the received information.

The determination unit 230 receives data on the inspection state or current state of the pump inverter, and compares it with the predicted result value predicted corresponding to the new inspection criterion generated by the inspection criterion generation unit 220, the You can check whether the inverter is in a normal state or a bad state.

In more detail, the inspection criterion generating unit 220 and the determining unit 230 may generate and determine the inspection criterion to maximize the performance of each pump inverter through the learned artificial neural network.

That is, by linking the information on the interceptor inspection for the pump and the individual pressure information of the pump stored in the inspection information storage unit 210 to create a data set that allows the pump to be optimized and maintained at a constant pressure, the data set is input and An artificial neural network through learning that outputs items related to results is applied.

The inspection reference generation unit 220 may construct effective inspection data through repeated inspection of the inverter for the pump.

For example, when the value of the output current to the value of the input current has a specific ratio, the corresponding inverter for the pump is the condition to optimally maintain a constant pressure in the pump. In this case, the frequency information and capacity information are Information through inverter inspection, such as what happens, can be matched with the pressure of the pump to learn the relationship. In this way, for each inverter for each pump, it is possible to present a set of inspection data on the optimal conditions for the pump to maintain pressure.

According to an embodiment, the inspection criterion generator 220 may generate a training data set and a verification data set for applying a deep learning algorithm. The data set may be generated by classifying the data set into a training data set required for artificial neural network learning and a verification data set for verifying learning progress information of the artificial neural network.

For example, the inspection standard generating unit 220 includes the state information to be randomly used in the learning data set and the verification data set among the state information of the pump inverter obtained from the inspection information storage unit 210 and the determination unit 230 . It is possible to classify the state information to be used for

In addition, the remaining data set selected for the verification data set may be used as the training data set, or a ratio of the state information of the pump inverter to be utilized in the verification data set and the state information to be utilized in the training data set may be set.

The inspection criterion generating unit 220 may generate a data set by dividing a data set for training and a data set for verification in order to prevent an overfitting state.

For example, since the training data set may be in an overfitting state due to the learning characteristics of the neural network structure, this can be prevented by using the validation data set. The validation data set may be, for example, a data set that does not overlap with the training data set.

Learning of artificial neural networks is through a backpropagation algorithm that applies the data set to a deep learning algorithm structure including a convolutional neural network and a fully connected deep neural network, and gradually changes the weight of the neural network structure corresponding to the error. It is characterized in that the results are fed back and learned.

That is, the determination unit 230 applies and learns a deep learning algorithm called a convolutional neural network to the data set built by the examination information storage unit 210 and the examination reference generation unit 220, and then pumps the data set. It can diagnose or predict whether the performance of the inverter is normal or bad.

The deep learning algorithm utilized by the determination unit 230 according to an embodiment may be composed of two parts: a structure of a convolutional neural network and a structure of a fully-connected neural network.

For example, a fully connected deep neural network forms a two-dimensional connection between nodes horizontally and vertically, so that there is no connection relationship between nodes located on the same layer, but only between nodes located in the immediately adjacent layer. can have

Since the details regarding the deep learning algorithm are obvious to those skilled in the art, a detailed description will be omitted below in order to further clarify the characteristics of the determination unit 230 according to an embodiment of the present application.

In addition, the determination unit 230 uses a convolutional neural network as an input to the training data set generated by the test criterion generator 220, and uses the output of the convolutional neural network as an input of a fully connected new layer neural network to learn a poor state. It is possible to build a model for judging

The determination unit 230 according to an embodiment compares the error between the result derived by applying the training data set to the deep learning algorithm structure and the actual result, and gradually changes the weight of the neural network structure as much as the corresponding error. The results can be fed back and learned through a backpropagation algorithm. Since the backpropagation algorithm is a well-known fact in the art, a detailed description thereof will be omitted.

Accordingly, the determination unit 230 may input the inverter status information for the pump acquired in real time to the artificial neural network built through learning to diagnose and predict the faulty status of the inverter for the pump.

), 펌프 설정 회전 수(

), 토출되는 유량(

), 토출되는 압력(

), 최대 유량(

), 최대 효율 유량(

), 인버터의 출력전압(

)을 입력데이터 삼아 펌프를 최적 효율로 운전 가능하게 하는 펌프용 인버터의 상태 기준들을 생성한다. In one embodiment, the inspection reference generation unit 220 according to an embodiment, for example, the set pressure (

), the number of revolutions set for the pump (

), discharged flow rate (

), discharge pressure (

), the maximum flow (

), the maximum efficiency flow (

), the output voltage of the inverter (

) as input data to generate the state standards of the inverter for the pump that enable the pump to operate with optimal efficiency.

)과 전압측정기(미도시)에 의하여 검출되는 직류전압(

)정보를 이용하여 출력전력(

)을 산출하는 단계, More specifically, the inspection reference generation unit 220, the output voltage of the input inverter (

) and DC voltage (not shown) detected by

) using the output power (

) to calculate

)와 출력전력(

) 및 토출되는 압력(

)을 이용하여 각 펌프의 토출유량(

)을 추정하는 단계,The set number of revolutions of the pump connected to each inverter for the pump (

) and output power (

) and discharged pressure (

) to the discharge flow rate of each pump (

) to estimate,

)과 토출되는 압력(

)에 따라 펌프의 회전수를 산출하는 단계,Inverter for pump is set pressure (

) and the discharged pressure (

) according to the step of calculating the number of revolutions of the pump,

)과 도 6의 펌프 시스템의 성능곡선 데이터에 기초하여 최대 유량(

)과 최대 효율 유량(

)을 추정하는 단계,Inverter for pump is set pressure (

) and the maximum flow rate (

) and the maximum efficiency flow (

) to estimate,

)을 수신하여 전체유량(

)을 산출하는 단계,Inverter for pump is the discharge flow rate of each pump estimated in the above step (

) to receive the total flow (

) to calculate

), 최대 유량(

), 전체유량(

), 펌프 설정 회전수(

), 설정압력(

), 토출되는 압력(

)을 이용하여 펌프가 일정 압력을 유지할 수 있는 최적 효율 운전점에서 구동할 수 있도록 하는 펌프의 필요 회전수를 산출 및 추정하는 단계,Maximum efficiency flow (

), the maximum flow (

), total flow (

), pump set rotation speed (

), set pressure (

), discharge pressure (

), calculating and estimating the required number of revolutions of the pump so that the pump can be driven at an optimal efficiency operating point that can maintain a constant pressure;

Finally, by weighting the performance curve data of the pump system and the output power of the pump inverter to determine whether the pump inverter can satisfy the required rotation speed of the pump calculated and estimated in the above step for a long time, the output power can be up to 22kW determining whether it does not exceed

go through

)은 하기 [수학식 1]에 따라 산출될 수 있다.For reference, the output power (

) can be calculated according to the following [Equation 1].

------- [수학식 1]

------- [Equation 1]

After that, when the determination unit 230 determines whether the state of the inverter for the pump to be inspected falls within the normal range, and the output power exceeds the maximum of 22 kW or is determined to be in a defective state, the inspection standard generating unit 220 determines the defective values. After analyzing and changing the voltage value input to the inverter for the pump by 1V unit through feedback, the above steps can be re-performed and a new test data reference can be generated.

The matters to be determined in the last step described above are not only whether the inverter for the pump, which is the inspection object, can satisfy the required rotational speed of the pump for a long time, that is, whether the pump can maintain a constant pressure for a long time, as well as the pump It is also possible to determine how quickly the set pressure can be reached, how well the constant pressure can be maintained despite the unexpected external pressure, and the like, but is not limited thereto.

More specifically, in order to determine how quickly the pump can reach the set pressure, in the step of calculating the rotation speed of the pump, not only the simple rotation speed but also the time to reach the calculated rotation speed of the pump are simultaneously measured. can judge In addition, in order to determine matters such as how well the constant pressure can be maintained despite the unexpected external pressure, it is determined that the above steps are performed in the same way after additionally inputting the atmospheric pressure value information around the pump in the input data. can

The inspection criterion generating unit 220 generates a new inspection criterion through feedback by the same algorithm even if the items to be determined are changed.

In addition, the input data input to the inspection reference generation unit 220 may include temperature information around the pump, humidity information, capacity information of the inverter for the pump, information on the power supply voltage, and the like.

For example, as the temperature increases, the movement of the fluid becomes more active and the discharge rate may be increased, and if the temperature is low, the movement rate of the fluid may decrease, so that the discharge rate may be slowed conversely. Therefore, the ambient or external temperature is divided into sections such as -10 degrees below zero, -10 degrees below zero, zero below 10 degrees, 10 degrees above zero, 20 degrees above zero, 20 degrees above zero and 30 degrees above zero. Inspection reference data sets can be created for each section.

)과 최대 유량(

) 및 전체유량(

)의 데이터값들이 보정없이 산출된다. 낮은 온도로 인해 유체의 운동성이 적어 유량 관련 데이터값들이 정상범주보다 낮은 값으로 산출될 수 있다. 이후 영하10도~영도 구간에서 재검사가 이뤄지는 경우, 해당하는 구간을 설정하여 입력하게 되면 검사기준생성부(220)는 이를 반영하여, 기준 온도인 영상20도~영상30도일때의 각 펌프의 토출유량(

)과 최대 유량(

) 및 전체유량(

)의 데이터값들과의 차이값을 비교한 후 해당하는 차이값만큼 보정하여 이후 단계를 거쳐 데이터를 산출하며 검사기준을 생성할 수 있다. 이로 인해 온도라는 변수에 의한 검사오류의 정도를 최소화할 수 있다.Specifically, first, the discharge flow rate (

) and the maximum flow (

) and total flow (

) are calculated without correction. Due to the low temperature, the fluid has low mobility, so flow-related data values may be calculated as lower than the normal range. Afterwards, when the re-inspection is performed in the range of -10 degrees to zero, when the corresponding section is set and entered, the inspection standard generating unit 220 reflects this, and discharges each pump when the reference temperature is 20 degrees to 30 degrees. flux(

) and the maximum flow (

) and total flow (

) after comparing the difference value with the data values, correcting the difference by the corresponding difference value, calculating the data through subsequent steps, and generating an inspection standard. Accordingly, it is possible to minimize the degree of inspection error caused by a variable called temperature.

The inspection information storage unit 210 according to an embodiment further includes a central server capable of storing and analyzing the result data derived from the inspection criterion generation unit 220 and the determination unit 230 based on the artificial neural network. can do.

The central server provides the operator with the analysis result data, such as the stage where the most errors occur, the configuration that the pump inverter has an error in, and the performance of the pump inverter recommended to solve the error or derive the optimal pump efficiency. This can increase the inspection efficiency.

Here, the central server can reduce coding time, can be easily developed for various platforms including mobile, and can be designed based on Microsoft's Azure PaaS, which can efficiently manage the life cycle of applications. . However, the platform used by the central server is not limited to Microsoft's Azure PaaS, and if it is a development environment that can play a role in reducing the defect rate of the pump inverter based on the artificial neural network proposed by the present invention, the name is not limited. All of them can be included.

The inverter inspection system 10 for a pump according to an embodiment may further include a notification unit 300 that receives a failure notification signal from the central server and automatically notifies an operator when an inverter for a defective pump is detected.

Referring to FIG. 7 , the lower end of the notification unit 300 may be formed in the shape of a square plate having a certain thickness, and the upper end may be formed in a cylindrical shape and a light diffusion unit may be provided along the outer surface, and the inner may be provided with an LED substrate.

The notification unit 300, when a failure notification signal is received from the central server, can emit light on the LED substrate, and uniformly radiate the light emitted from the LED substrate in all directions through the light diffusion unit provided along the outer surface. have.

Referring to FIG. 5, the inspection sequence of the finished product inspection unit 200 will be described in detail.

First, when the inspection execution signal is input, the set value corresponding to the inspection execution signal is transmitted to the pump inverter and the inspection information storage unit 210 . At this time, the inspection information storage unit 210 stores a data set based on the set value, and the inspection reference generation unit 220 generates an optimal state data set of the inverter for the pump that allows the pump to be maintained at a constant pressure. After that, it outputs the state measurement and state data of the inverter for the pump that occurs as the pump operates. Thereafter, the determination unit 230 inspects the state of the inverter by comparing the result value with the data set generated by the inspection reference generation unit 220 . The determination unit 230 compares the two to determine whether the pump inverter outputs a normal value or a bad value, and determines whether a normal value (eg, input current or voltage and output current and voltage match, etc.) value), the test ends. In the case of outputting a defective value, the above-described artificial neural network analyzes and merges the defective data, and provides feedback on the optimal data set information to the inspection standard generating unit 220 and the determining unit 230 so that the inspection is resumed. do. With the algorithm as described above, it is possible to find the optimal conditions for each of the inverters for the pump that can maintain a constant pressure of the pump, and automatically select the inverter for the pump that satisfies such conditions through this inspection system.

In the above, preferred embodiments of the present invention have been illustrated and described, but the present invention is not limited to the specific embodiments described above, and it is common in the technical field to which the present invention pertains without departing from the gist of the present invention as claimed in the claims. Various modifications are possible by those having the knowledge of, of course, these modifications should not be individually understood from the technical spirit or prospect of the present invention.

10. Inverter inspection system for pumps
100. Board stage inspection department
200. Finished product inspection department

Claims (2)

a board stage inspection unit that inspects whether each board operates normally at the board level when the inverter for the pump is manufactured; and
and a finished product inspection unit that detects a product error in the finished product assembly stage of the pump inverter after the inspection by the board stage inspection unit;
The board step inspection unit,
a test jig unit that turns on a corresponding switch for signal detection according to the switch control signal, and detects a test signal from the board to be measured through the turned on switch;
a frequency counter unit for counting the frequency of the detection signal output from the test jig unit;
a signal input/output unit for inputting and outputting signals for controlling switches for signal detection of the test jig unit;
a signal processing unit converting the test signal detected by the test jig unit into digital data;
a main control unit that processes data input from the signal processing unit and the frequency counter unit by inspection/adjustment software and outputs a switch control signal of the test jig unit according to the result to the signal input/output unit; and
a display unit for outputting an adjustment item selection signal and an adjustment reference value to the main control unit, and receiving and displaying display data from the main control unit;
The signal processing unit,
a first selection signal output unit for selecting and outputting one of the two signals detected by the test jig unit and inputted from the first channel and the second channel;
a first data conversion unit converting the signal output from the first selection signal output unit into data;
a sub-control unit processing the signal data output from the first data conversion unit by inspection/adjustment software and outputting character data according to the result;
a signal processing unit receiving the synchronization signal output from the first data conversion unit and outputting an enable signal;
an address counter receiving the enable signal output from the signal processing unit and outputting address data;
a memory unit receiving and storing character data from the sub-control unit, receiving address data output from the address counter, and outputting data stored in the address;
a second data conversion unit for outputting the data output from the memory unit as an analog signal; and
a second selection signal output unit for selecting one of the first selection signal output unit and the second data conversion unit and outputting an input signal to a display;
The finished product inspection unit,
an inspection information storage unit storing inspection steps and inspection standards for the performance inspection of the inverter for the pump;
an inspection standard generation unit for receiving an inspection step and an inspection criterion matching the inspection execution signal from the inspection information storage unit, and analyzing the received information to generate a new criterion; and
A determination unit that receives the state data of the inverter for the pump and compares it with a predicted result value predicted in response to the new inspection criterion generated by the inspection criterion generation unit to determine whether the inverter is in a normal state or a bad state; including,
The inspection standard generating unit and the determining unit,
Create a data set that allows the pump to be optimally maintained at a constant pressure by linking the information on the inspection of the inverter for the pump and the pressure information of the pump stored in the inspection information storage unit,
It is characterized in that applying an artificial neural network through learning that takes the data set as an input and outputs an item related to the result,
Learning of the artificial neural network,
applying the data set to a deep learning algorithm structure including a convolutional neural network and a fully connected deep neural network;
It is characterized in that the result is fed back and learned through a backpropagation algorithm that gradually changes the weight of the neural network structure corresponding to the error,
The inspection standard generating unit,
The set pressure of the pump system to which the inverter for the pump is connected, the set number of rotations of the pump, the discharged flow rate, the discharged pressure, the maximum flow rate, the maximum efficiency flow rate, the input voltage and current of the inverter, and the output voltage of the inverter can be used as input data,
Detected by the input inverter output voltage and voltage measuring device? Calculating the output power using the DC voltage information;
estimating the discharge flow rate of each pump using the set rotational speed, output power, and discharged pressure of the pump connected to each of the pump inverters;
calculating the number of rotations of the pump according to the set pressure and the discharged pressure of the pump system, by the inverter for the pump;
The step of the inverter for the pump estimating the maximum flow rate and the maximum efficiency flow rate based on the set pressure of the pump system and the performance curve data of the pump system;
The inverter for the pump receives the estimated discharge flow rate of each pump and calculates the total flow rate;
Using the maximum efficiency flow rate, maximum flow rate, total flow rate, set pump rotation speed, set pressure, and discharge pressure, calculate the required number of revolutions of the pump so that the pump can be operated at the optimum efficiency operating point that can maintain a constant pressure and estimating, and
Whether the pump inverter can satisfy the calculated and estimated required rotation speed of the pump for a long time is weighted on the performance curve data of the pump system and the output power of the pump inverter to determine whether the output power does not exceed the maximum of 22kW characterized in that it comprises the step of
After that, if it is determined by the determination unit whether the state of the inverter for the pump to be inspected falls within the normal range, and the output power exceeds the maximum of 22 kW or it is determined to be in a defective state, the inspection standard generating unit analyzes the defective values and provides feedback to the pump. Inverter inspection system for pump, characterized in that by changing the voltage value input to the inverter for 1V unit, and generating a new inspection data standard after each step is repeated.
delete

Images