Best table/index structure for data with time-varying association - mysql

Question

My company uses a number of data acquisition devices (DAQs) to monitor the output of solar panels on a test site. Each DAQ has a unique serial number, and each solar panel has a unique serial number. Occasionally we swap out the panels for new panels, and occasionally the DAQs fail and need to be replaced with new ones with different serial numbers.
My question is, what is the best table structure for queries to see all of the data for a particular solar panel's serial number, given that it can be on different DAQs at different times?
I'm currently using the following table structure:
Table: relationships
id int(11) NOT NULL AUTO_INCREMENT,
daqID char(4) NOT NULL,
dtdateFirst datetime NOT NULL,
dtdateLast datetime NOT NULL,
PanelType varchar(20) NOT NULL,
sgcucode varchar(45) NOT NULL,
serial varchar(15) NOT NULL,
ptype varchar(15) NOT NULL,
PRIMARY KEY (id),
KEY daqID (daqID),
KEY gcuidx (sgcucode),
KEY serialidx (serial),
KEY fullidx (sgcucode,daqID,serial,dtdateFirst,dtdateLast)
) ENGINE=InnoDB AUTO_INCREMENT=135 DEFAULT CHARSET=utf8
Table: data
id int(11) NOT NULL AUTO_INCREMENT,
dtdate datetime NOT NULL,
daqID char(4) NOT NULL,
Varray text NOT NULL,
Iarray text NOT NULL,
Iavg float NOT NULL,
Pmp float NOT NULL,
PRIMARY KEY (id),
UNIQUE KEY id_UNIQUE (id),
UNIQUE KEY dupliData (dtdate,daqID,Iavg),
KEY idxDaqDate (daqID,dtdate),
KEY idxDate (dtdate),
KEY idxPmp (Pmp)
) ENGINE=InnoDB AUTO_INCREMENT=14027571 DEFAULT CHARSET=utf8
The "relationships" table matches the daqID to a panel serial number for a given time span (from "dtdateFirst" to "dtdateLast"). So the important columns for this question are: daqID, dtdateFirst, dtdateLast, serial. Also of some importance is the "sgcucode". This column indicates which test site the modules are on. It is used by a dashboard so that we can cycle through various sites which log data to the same table.
Data is constantly being logged to the "data" table from the DAQ devices. I then use the relationships table to correlate the serial number of the solar panel with the correct daqID for the time in question.
By far the most common query is to collect all of the data in the "data" table for a given day, and display it in a dashboard (shown below).
This is the query I use to do this:
SELECT relationships.serial as title, dtdate as time , Pmp as Value, relationships.ptype as type
FROM data INNER JOIN (relationships) ON (relationships.daqID=data.daqID)
AND dtdate BETWEEN DATE_FORMAT(example_date, '%Y-%m-%d 05:00:00') AND DATE_FORMAT(example_date, '%Y-%m-%d 21:00:00')
WHERE relationships.dtdateFirst <= dtdate
AND relationships.dtdateLast >= dtdate
AND sgcucode="example_code";
Given these conditions, is this the best solution? I probably have redundant indexes, I am still learning about database design, so any suggestions for improvement would be greatly appreciated!